(navigation image)
Home American Libraries | Canadian Libraries | Universal Library | Community Texts | Project Gutenberg | Children's Library | Biodiversity Heritage Library | Additional Collections
Search: Advanced Search
Anonymous User (login or join us)
See other formats

Full text of "Notes on the nebular theory : in relation to stellar, solar, planetary, cometary and geological phenomena"














F.E.A.S., F.G.S., RE.MET.SOC., M.PnYs.Soc., 


" A cause qu'il ne convient pas si bien a la souveraine perfection qui est en 
Dieu de le faire auteur de la confusion que de 1'ordre, et aussi que la notion 
que nous en avons est moins distincte, j'ay cru devoir icy preferer la proportion 
et 1'ordre a la confusion du Chaos." Descartes. 







THE subject of the following pages as a study commenced 
to fascinate me in my youth, when, deeply imbued with 
the study of Newton, I was trying vainly to unravel the 
theoretical possibilities of the original structure of the 
marvellous mechanism of the universe. It has been ever 
present with me throughout a busy life and has been 
often reconsidered, leading me to the conclusion that a 
modified form of the Nebular Theory of Laplace might 
be established on some new ideas which I formed and by 
certain calculations that I felt sure the actual conditions 
warranted. These speculations are now offered to the 
scientific world for approval. 

I arranged this matter as it presented itself to my mind 
originally in papers upon separate parts of the subject, as 
a less confident mode of introduction ; but I was advised by 
orthodox authorities that such papers were too speculative 
to communicate to the learned societies that I thought 
at the time best adapted for their consideration. These 

papers, which I now edit in abstract, have been put aside 




for many years. Upon the permissible borders of the subject 
I read a paper before the Geologists' Association in March 
1883, " Upon the Causes of the Elevation and Depression 
of the Earth's Surface," which I considered to represent 
the continuity of effects of nebular conditions upon the 
Earth (reported in l Nature,' March 29th, 1883). Some 
of the speculative unpublished matter of this paper is in- 
cluded in the following pages, with copies of my diagrams. 

I wrote a paper in 1878 upon " Some Hypothetical 
Conditions of the Properties and Motions of Comets," that 
I assume to depend upon original nebular conditions, which, 
after some correspondence with a high authority, I sent to 
the l English Mechanic/ a journal in which astronomical 
subjects are frequently discussed (published 22nd June, 
1883). This matter is incorporated herein with some con- 
clusions arrived at after further consideration of the subject. 

At the British Association in 1883 I read a paper entitled 
" Notes upon the Rotation-period of the Earth and Revo- 
lution-period of the Moon, deduced from the Nebular 
Hypothesis of Laplace" (Brit. Assoc. Reports, 1885, p. 915). 
The subject of this paper is more fully treated in the 
present work. 

I also read a paper before the Geological Society showing 
the error in Mallet's theory of the contraction of the Earth 
in cooling, which process was assumed to produce the 


great elevation and inclination of strata observed in nature 
(Phil. Trans, vol. clxiii. part 1). This theory, which has 
been very generally accepted, derived most of its support, 
I think, from a large accidental error which I was able to 
point out in the mathematical calculations (Proc. Geol. Soc. 
June 1884), so that the causes of elevation and inclination of 
strata discussed herein as a process of the continuity of 
nebular conditions may be said to remain unexplained by 

any hypothesis founded upon correct data. 
'J ifcriffi yyjj.t [jjnigno w> 80bi *j; nolhoqtnq r?i ( 3ds(tf#fi 

I have also read several papers before the British Asso- 
ciation and the Geological Society contravening some points 
in the popular theory of a universal glacial age, which I 
think was only local at any period, and is opposed to the 
Nebular Theory, which I thought, at the time I wrote the 
papers, demanded a uniform decrement of heat in time 
in the entire cosmic system. This idea as regards the 
periodic amount of solar radiation to the Earth I have 
somewhat modified in these pages by considering the effects 
of critical temperatures upon the solar nebula ; but as I 
have endeavoured to bring the whole subject together as 
briefly as possible as it presents itself to my mind, it is 
unnecessary to discuss more particularly what I have 
already attempted to do in this direction. The greater 
part of this matter has remained unpublished, except partly 
in short abstracts, being at present somewhat out of concord 
with prevailing theories. 


What 1 most regret in this matter is that I am unable 
to fully discuss many theoretical ideas that have been pro- 
posed by scientific men without extending this sketch very 
much beyond the limit necessary for the brief statement of 
my own ideas. This is unfortunate, as I find in reading 
up the subject, for the most part since I wrote these pages, 
that other ideas approach my own in a few particulars. 

One feels also that, in the discussion of a speculative 
subject, in proportion as ideas are original they must be 
difficult to correlate with the more or less established scientific 
theories. The whole subject, however, is undoubtedly in a 
tentative state, and must be studied generally upon a broader 
and more exact basis in detail than it has heretofore been, 
if a satisfactory theory is to be established. For this inves- 
tigation some acceptable data must be found before refined 
mathematical analysis can be of much value. For this 
theory I can only hope I have put forward some available 

I am indebted to Mr. W. T. Lynn, B.A., F.R.A.S., for 
critical examination of my proofs both for reading and for 
calculations. I am indebted to the kindness of Mr. William 
Crookes, F.R.S., and Professor G. F. Fitzgerald, F.R.S., 
for some suggestions which have enabled me to render the 
matter in the second and twelfth chapters more definite and 
logical. I am indebted to an eminent practical geologist, 
who withholds his name, for reading and suggestion in 


Chapters X. to XVI. ; to Mr. Charles Kirk for care in 
the reproduction of the Plates ; and to Mr. W. Francis, of 
my printers' firm, for care in correcting my proofs. 

South Norwood, March 1895. 

floi:foo?.sKI s' 

.q ^.o 

BoiIoO .OP. .q , nia^y8 cmirrm 1 ! u ni uoiioA 

K IfloimonoiteA r io aoLtonoBdO *&: .q 

.Oo . ,nu8 ?,'iii !io rroliibaoO 


iunnol : 

.feli .Q ( fned8-( jjonrsn jini^o oi moil G&rjdla o 
,/5fnj;on c I lunrgho Qiia nr enoiiibnoO avtooM J^aegSffS 


v-H^flui^-^Joa M i - 1Ari ^ t ^ ;I fJ doK l jrikl a 10 

naissance, p. 2. Descartes Wright, p. 3. Kant, p. 4. 
Lambert Sir Wm. Herschel, p. 6. Laplace, p. 7. 
Mayer, p. 9. Helmholtz, p. 10. Lane Faye, p. 11. 
Projectile Theory Limits of Theory to be discussed, 
p. 15. 

\Tl tt Tl T^T t TT *'\ 





Time and Space, p. 17. Original State of Matter, p. 18. 
Separation of Systems of Original Matter, p. 19. Com- 
parison with Astronomical Nebulae, p. 20. Tenuity of 
Original Matter, p. 21. Milky Way Atomic Theory, 
p. 22. Sir Benj. Brodie's Theory of Original Matter, 
p. 23. Suggestions for the Constitution of the Nebulae 

Pneuma, p. 23. Lockyer's Dissociation Crookes's 
Fractionation, p. 25. Pneumites, p. 26. Chemical 
Action in a Pneuma System, p. 30. Cohesion of Matter, 
p. 32. Observation of Astronomical Nebulae, p. 35. 
Condition of the Sun, p. 36. 


of Nebulae from the original Pneuma System, p. 38. 
Suggested Motive Conditions in the original Pneuma, 
p. 40. Cyclone-inducing Conditions, p. 42. Formation 
of Spiral Nebular Systems, p. 45. Solar-Planetary 
inducing conditions in Nebulae, p. 46. Permanency of 
Stellar Systems, p. 47. Distances under which Gravita- 
tion may be active Action of Gravity in formation of 
Circular Orbits, p. 49. 


Separate Stellar and Solar- Planetary inducing con- 
ditions according to amount of Rotation, p. 52. Limits 
of a Solar-Planetary-Cometary System, p. 53. Action 
of Gravity on distant Condensations, p. 55. Condensation 
to a Solar Centre, p. 56. Direction of Approach to the 
Sun of exterior Matter Formation of Orbits, p. 58. 
Cometary Orbits, p. 60. Formation of a Planetary 
Plane, p. 62. Planets formed at the Perihelion of 
Cometary Orbits, p. 63. 


Energy of the Solar System Its Asymmetry, p. 65. 
Scheme of a Symmetrical Gaseous Solar-Planetary 
System, p. 67. Modes of Condensation of Interior 
Planets in a Spheroidal Nebular System, p. 71. Mode 
of Condensation of Extreme Outer Solar Nebula, 
p. 72. Breaking-up of Gaseous Zone-Systems, p. 73. 
Influencing Conditions of Planet-formation Critical 
Temperatures, p. 74. Some Modifying Conditions, p. 77. 


tances of the Planets from the Sun Bode's Law, p. 80. 
Masses of the Planets, p. 81. Proportional Densities, 
p. 82. Probable Form of the Original Planetary Nebula, 
p. 84. Effects of the voluminous Ring of Jupiter upon 
the Intra-Jupiter Solar System, p. 85. Relative Rate of 
Cooling of the Intra-Jupiter System with the Earth, 
p. 86. 



SATELLITES : Direction of the Solar Axis of Rotation, 
p. 88. Direction of the Planets' Axes, p. 89. Rotation 
of the Sun, p. 90. Momentum of a Planet, p. 95. 
Orbital Velocity of a Planet, p. 96. Rotation of Planets, 
p. 97. Calculated Rotation of Jupiter and Saturn, 
p. 101. State of Jupiter and Saturn, p. 103. Rotation 
of the Asteroids of Mars, p. 106 . Rotation of the Earth, 
p. 108. 




Revolution of Satellites with Direct Motion, p. 110. 
Calculation of Revolution of the Satellites of Jupiter and 
Saturn, p. 111. Satellites of Mars, p. 113. The Moon, 
p. 114. Retrograde Motion of Satellites, p. 116. 


SYSTEM : Discussion of Principles, p. 121. Comets of 
Long Period, p. 125. Comets of Short Period Sym- 
metrical Elements of Comet-formation, p. 126, Comets 
considered as Gravitative Matter, p. 128. Conditions 


under which a Comet may be considered as a Planetary 
Body, p. 130. Heat Electricity Orbital Momentum, 
p. 131. Elongation of the Cometary Mass near Peri- 
helion, p. 136. Orbits of the outer parts of a Comet 
Focal Point, p. 139. Direction of a Cometary Train in 
relation to the Sun, p. 140. Widening Curvature of the 
Train by crossing Orbits, 142. Formation of a New 
Head to a Comet, p. 144. Some general conditions, 

iji>o{o(I 'io Lobt/i .081 .q r gj;*m;-Lflj)J lo 


sidered as a Model Planet, p. 147. Factors of Earth- 
formation, p. 148. From a Gaseous System, p. 149. 
From aggregation of Planetoids, p. 150. Internal 
Fluidity of the Earth discussed, p. 150. Effects of 
Tidal Friction, p. 156. Change of figure due to Rota- 

tion-Velocity, p. 157. 
J ' F 


Y3L (ISO "KlDm KOIT&MJIOH-HTH.A2 "50 XOiTIdXoD OTfTOrfl'i^f 

mation of Land-areas by Inclusion of Planetoids, p. 160. 
Projection of a large Intra-Mars Planetoid upon the 
Earth, p. 165. Formation of a Continent therefrom, 
.81 P- 1"6. 



PURELY NEBULAR CONDITIONS : Conditions specialized, 
p. 169. Distinct Periods of Deposition Period of 
Condensation of highly Refractory Matter, p. 170. 
Period of Condensation of Volatile Metals, Oxygen, and 
Halogens with Metals and Metalloids, p. 174. Distri- 
bution of Land-areas, p. 180. Period of Deposition of 
Water, p. 184. 


ICE AT THE POLES : Pre-Glacial and Glacial Period s ; 
p. 189. -Distribution of Ice at the Poles, p. 192. 
Present Conditions brought about by Deposition of Ice, 
p. 193. Where Ice-pressures are most active, p. 197. 
Extrusion of Volcanic Matter through Ice-pressures, 
p. 199. Pressure of Water and iSteam in Volcanic 
Eruptions, p. 201. Notes upon Theories proposed 
p. 207. 


Condition of the Sun during Earth-formation, p. 210. 
Distinct Solar-heating Periods, p. 213. Division of 
Special Periods, p. 214. General effects of the Large 
Nebulous Sun upon Meteorological Conditions, p. 218. 



sation of the Solar System, p. 220. Calculation of Time 
of Condensation for the Nebula extending to the Orbit of 
Neptune, p. 222. The same for the Orbit of the Earth, 
p. 223. Distribution of Time upon the Earth, through- 
out the Varying Periods of Condensation of the Sun 
and the Inferior Planets, p. 224. Table in millions of 
years Period of the Formation of the Nebulous Earth, 
p. 225. 


NOMENA : Discussion of Geological Periods, p. 228. 
Archaean Period, p. 231. Archaean Time in Relation 
to the Conditions of Animal Life, p. 233. Cambro- 
Silurian Period, p. 235. Devonian Period, p. 239. 
Permian Period, p. 241. Triassic and Rhaetic Periods 
Jurassic Period Cretaceous to Tertiary Periods, p. 243. 
Glacial Period, p. 246, Future Period, p. 249. 

APPENDICES. A. Hypothesis of Radiation of Heat and Light, 
p. 252. B. Mallet's Theory, p. 256. C. Contempo- 
raneous Stratification of Rocks of the prevailing Chemical 
Elements, p. 258. 






THE origin of the material universe has occupied the deepest 
thoughts of many of our most profound thinkers. These 
speculations must, however, rest for ever upon the borderland 
of Science, where few practical men may care to tread. To 
bring this subject more nearly into co-relation with practical 
science it is necessary that we keep more nearly to the 
evidences of natural phenomena for our data than has here- 
tofore been done by our scientists, who have so often in this 
matter followed pure speculations only. We may also 
possibly with advantage make tests of the theories that have 
been proposed by submitting them to calculations taken 
directly from the premises offered in the various hypotheses. 
To place this matter before the reader clearly for discussion, 
it will be convenient to offer a few historical notes on the 
leading theories that have been proposed, upon which it is 
my purpose to graft certain suggestions and to make certain 



calculations. These historical outlines will save space being 
taken by constant definitions of the earlier theories, by making 
use of references to the paragraph numbers to be found in 
this chapter. 

1. The ancient ideas of the Kosmos in no way approach 
the possible conditions of a Nebular theory the extent of 
the universe in early times was conceived to be only that which 
appeared evident to the senses, the earth being taken as the 
centre of the universe, surrounded and enclosed by firma- 
ments that were assumed to be revolving solid constructions, 
which were variously defined. To this, however, we have 
some exceptions. Anaximines believed stars to be of fiery 
substance and to carry invisible earthly bodies with them. 
As a preliminary idea of early nebular conditions he held 
that air was the original material of the universe, from which 
all things were engendered and into which they resolve *. 
Pythagoras taught his disciples that the sun was the centre 
about which the planets revolved f, by which he accounted 
for eclipses and the motions of the planets, and thereby 
clearly anticipated what we term the Solar System of Coper- 
nicus. This was a great advance upon the prevailing theory, 
which was limited to a firmament or a number of crystal 
spheres surrounding the earth. The theory of Pythagoras, 
probably derived from the Chaldeans, was, however, far too 
much in advance of the age to be accepted by the following 
generations of popular scholars and theorists, who were, in 
many instances, strongly prejudiced against it owing to the 
influence of the prevailing superstitions of the time, in 
accordance with which alone popularity could be attained. 

2. The earliest suggestion of a nebular hypothesis that 
occurs within the Kenaissance period is probably that of 
Tycho Brahe, who, to account for the new star which appeared 
in 1572, suggested that stars were formed by condensation of 

* ' History of Philosophy,' T. Stanley, 4th e<L 1743, p. 54. 
t Id. p. 444. 


the ethereal substance of which he imagined the Milky Way 
was composed *. Kepler accepted Tycho Brahe's idea, which 
he somewhat extended in his account of the new star which 
appeared in 1604, by suggesting that the nebular substance 
might not be confined to the Milky Way alone but may have 
pervaded all space |. In an account of the eclipse of the sun 
at Naples in 1605 he suggests that nebular matter is of the 
same kind as that which appears around the dark body of the 
moon in a total eclipse. 

3. Descartes appears to have been the first to attempt to 
construct a complete theory of the origin of the known 
universe, or to systematize the matter of space. In this he 
assumes that universal matter originally existed in three 
states coarse, fine, and very attenuated ; that it drifted 
originally in a complex system of whirls, and that each of 
these whirls formed a solar or planetary system J. This 
theory, brought forward again recently by M. Faye, will be 
presently discussed : we may term it the Vortex Theory. 

4. Thos. Wright was the first to suggest a complete gravi- 
tation theory of the universe founded upon astronomical 
observations taken accurately enough to be of any scientific 
value . He suggests that the universe, represented by the 
Milky Way, is a unit gravitation system in general revolu- 
tion in the form of a bifurcating stratum composed of all the 
visible stars, which resemble our sun. This is now commonly 
termed the Grindstone Theory. He estimated the direction 
in which our sun is travelling among the stars by discussion 
of the parallax. He suggests that the stars, by uniformity 
of creation, have revolutionary subsidiary systems of planets 
similar to our own solar system. In the plate which forms 

* ' Progymnasmata,' 1572, p. 795. 
t l Stella nova in pede Serpentarii,' 1606, p. 115. 
| ' Essais Philosophiques,' 1637. l Speciminse Philosophic^,' 1644. 
' An Original Theory of the Universe/ Thomas Wright, 1750. See 
also De Morgan, Phil. Mag. 3 ser. xxxii. p. 241 . 



our frontispiece he shows by shading the theoretical gravita- 
tion influences over matter of our Sun, Sirius, and Higel, in 
which each of these star systems is shown extending its 
influence over an approximately equal area of space, and 
nearly meeting the system of each of the others. He 
suggests that with more perfect telescopes the rings of 
Saturn will be discovered to be formed of small satellites. 
He considers the measurable visible sun to consist of a 
vaporous and a nebulous atmosphere, the dense solid or 
liquid body of the sun being much smaller than it appears, 
possibly only of about two thirds its apparent dimensions. 
He contends that the Milky Way forms one vast system 
composed of solar systems like our own. He does not appear 
to know of more than the six nebulse mentioned by Halley as 
"light coming from an extraordinary large space in the ether, 
through which a lucid medium is diffused, which shines with 
its own proper lustre"*. Wright refers to these cloudy 
spots as " condensations of vapour among the mass of stars 
to which our sun belongs." Comets are suggested to have 
elliptical closed orbits, as represented in the frontispiece, and 
therefore to be periodical. 

5. Wright's original bold but (as regards particulars left 
unnoticed) somewhat indefinite outline was filled up more in 
detail by Kant, who fully recognizes the speculations of 
" Wright of Durham/' and accepts his general principles 
concerning the structure of the universe. Kant's original 
speculations given in the second part of his work are princi- 
pally directed to account for the formation of the solar 
system, the mass and motion of which are assumed to have 
been produced by the aggregation of free particles that 
were formerly uniformly distributed in space in an attenuated 
form. The particles falling together at an early period by 
initial gravitation formed masses by local condensations. 

* Halley, Phil. Trans. 1714. 


These masses, under universal gravitation, are assumed to 
have encountered the resistances of other masses and particles, 
generally distributed, so that those parts of the system only 
could continue to move freely and form concentric systems 
which acquired a linear velocity sufficiently in equation with 
the nearest centralizing attraction to produce orbital motion. 
The matter deflected from the direct line of attraction towards 
the sun passed into revolution about it. The revolution pro- 
duced a denser extended equatorial plane, into which exterior 
matter was drawn whilst approaching the sun. The velocity 
of the particle falling towards the sun depended upon the 
distance fallen, the direction it finally took upon the sum of 
lateral deviations it experienced in consequence of encounters 
with other particles, which directed it, under the influence of 
gravitation, into the path of least resistance. The particles 
which did not meet the conditions of circular or orbital 
motion fell into the sun, where his attraction predominated. 
The particles deflected or held in equilibrium in nearly the 
same orbit formed the planet by gravitating towards denser 
condensations of surrounding matter, which, acting under 
similar conditions, took the same direction of revolution as 
the sun. The same motive principles in matter which pro- 
duced the revolution of the planet around the sun also 
produced its own rotation and the revolution of its satellites 
in the same direction. The planet in regard to rotation is 
considered as an independent body. The rotative movement 
might therefore, according to the momentum of the mean 
drift of its matter, take one direction or the other, the 
direction of rotation being due to the unequal velocity of the 
particles in circulation around the sun at the time they were 
falling upon the new forming planet through its prevailing 
local attraction. Saturn is taken as a particular case for 
consideration, in which vaporous conditions of condensation 
are suggested, the ring being in revolution and thrown off 
the planet where the centrifugal force of its matter becomes 


in equilibrium with gravitation *. It will be convenient to 
denominate the nebular theory of Kant the Discrete System, 
in contradistinction to the Concrete or gaseous system of Sir 
William Herschel and Laplace, notices of which follow. 

It is readily seen that matter equally distributed in space 
could not possibly drift in the manner proposed by Kant, but 
that it must at an early period fall into the whirl system of 
motion proposed by Descartes. M. Faye, although generally 
supporting the discrete theory of Kant, has demonstrated 
that if matter drifted under the influence of gravitation only, 
as proposed by this philosopher, it would possess no rotation 
upon condensation in forming the sun or a separate planetary 
system t 

Lambert followed closely in the theory of Kant, his greatest 
divergence being in the division of the universe into many 
galactic systems of which our Milky Way represents one 
only f. This is now denominated the Island Theory. 

6. No further advance was made in the nebular theory 
until over 2000 nebulae had been discovered and examined 
by Sir William Herschel, an account of which was placed 
before the Koyal Society in several papers from 1784 on- 
wards. The nebulae were recognized individually as immense 
gravitation systems in 1789. The planetary nebulae were 
adduced as giving evidences of atmospheres of shining fluid 
about stellar foci, which were suggested to be in a state of 
condensation in 1791. The entire subject is brought together, 
embracing ideas of the origin of our own solar system being 
derived from nebular matter, in Phil. Trans. 1811, p. 269 et 
seq. The conclusions arrived at are somewhat less original 
than Herschel supposed. They possibly mark in one respect 
the influence of chemical discovery, in which the smallest 
parts of bodies were beginning to be recognized as distinctly 

* ' Allgemeine Naturgeschichte und Theorie des Himmels,' 1755. 
f ' Sur rOrigine du Monde/ 2nd ed. p. 135. 
J l Cosmologische Briefe, ' 1761. 


structural units *, all of which might be brought to a gaseous 
state by heat : therefore the possibility of the existence of all 
known matter in three forms gaseous, liquid, and solid 
dependent upon the special temperature to which any element 
is exposed. The recognition of a possible gaseous state for 
all matter appears to have suppressed for the time the former 
prevailing idea, as regards the nebular hypothesis, that 
material particles separately distributed in space represented 
the most attenuated form of matter. The gaseous element 
offered also at the same time a new foundation for the con- 
struction of a nebular hypothesis. 

7. The subject is taken up by the powerful analytical 
mind of Laplace in 1796 f and in 1799, and advanced in 
following years J. It is treated entirely de novo, this author 
evidently not knowing how much had been thought out in 
the same direction by others. He follows without knowing 
it closely upon the general arguments of Kant, particularly 
those of his theory of the formation of the rings of Saturn, 
with the important addition of the introduction of the gaseous 
nebular element as a universal medium. He formulates that 
whatever could have directed the movements of the planets, 
it must have been originally a concrete system embracing 
the whole of these bodies, which could not possibly, seeing 
its immense extent, have been other than an aerial fluid 
surrounding the sun and possessing the same direction of 
revolution. To ensure this possible extension of matter he 
supposes that the nebulous gaseous matter was of sufficiently 
high temperature for all solids to exist in a purely gaseous 
state. He suggests that this attenuated matter of the solar 
system probably resembled some of the nebulae visible in the 
telescope, or more particularly the nebulous stars which were 
formed from the general more attenuated and highly heated 

* Lavoisier's ' Traite* elSmentaire de Chemie,' 1789. 
t * Exposition du Systeme du Monde/ vol. ii. p. 295. 
J Id. 3rd edit. 1813 


nebulous matter. The original momentum of the solar 
nebula in revolution was conserved, so that as it contracted 
it attained higher velocity. The planets were formed at the 
limits of the solar atmosphere when it was a planetary nebula 
by successive zones of vapour being abandoned in the plane 
of the sun's nebular equator, at a radius where the centri- 
fugal force of the zone, due to its contraction and accelerated 
rotation, was in equation with gravity for the orbit of the 
planet. After separation the parts of the zone-ring would 
maintain the same angular velocity as they had while in 
contact with the sun for a time ; but in falling towards the 
new planet forming by local condensation, the exterior 
matter, besides its excess of linear velocity over interior 
parts due to its exterior position, would attain further excess 
of velocity through gravity. The downward impulse of 
gravitation into tangential velocity would impress an excess 
of velocity over the original angular velocity in condensation 
upon the planet, and thereby cause its rotation to be in the 
same direction as the revolution of the planet around the sun. 
This effect will be discussed, with a diagram, in the body of 
the work. The satellites were formed at the limits of the 
atmosphere of the nebulous planets at an early period in the 
same manner as the planets themselves were formed about 
the sun. He suggests that comets are of another system 
with linked orbits, and that they move under the mutual 
attractions of our sun and other separate stars. The theory 
of Laplace has many able supporters, among the most able of 
whom is the celebrated astronomer M. C. "Wolf*, who has 
made important additions to it. 

8. The discovery of certain factors of the mechanical 
theory of heat by Mayer in 1842 led this philosopher in 
1848 to propound a theory of the possibility of the sun's heat 
being maintained by the percussion of the fall of meteoric 

* ' Les Hypotheses Cosmogoniques,' 1886, p. 35. 


matter upon his surface *, a principle entailing the formation 
of the sun from such matter, or what became the " Meteoric 
theory of the Sun" This theory, which resembles that of 
Buffon, wherein the sun's heat is suggested to be maintained 
by the fall of comets constantly upon his surface, was ex- 
tended to the formation of the planets also. It was thought 
to be supported by the resolution of many presupposed 
gaseous nebulas into stars by means of the great reflecting 
telescope of Lord Kosse, erected in 1845. These observa- 
tions led many astronomers to hold that all nebulas would be 
resolvable if sufficient optical power were applied to them, 
and, therefore, that we possess no certain inference of the 
presence of incandescent gaseous matter in space. 

9. Mayer's theory was fully investigated by Lord Kelvin, 
and its entire insufficiency to account for the dispensation of 
solar heat clearly demonstrated f- It was supported by 
Prof. Tait, who extended the conditions by suggesting that 
the incandescent state of nebulae which were afterwards 
spectroscopically demonstrated as being gaseous matter by 
Dr. Huggins, might be derived from collisions of small 
cosmical bodies which were surrounded by gaseous atmo- 
spheres. Lord Kelvin, in reconsidering the meteoric theory 
with this modification, thought that, although the sun might 
have been formed by the cohesion of small masses, his heat 
could not be maintained by the mechanical collisions of 
gravitating matter only. He considers the low rate of 
cooling, and the consequent constancy of emission of heat, 
as being covered in great part by the high specific heat of 
the matter of the sun. He shows that if the earth fell 
directly to the sun, this would only maintain its present heat 
for another 95 years J. He states that if the whole mass of 
the planets were to fall into the sun from their orbits, the 

* 'Dynamik des Himmels,' 1842, p. 12. 

t Trans. R. S. Edinburgh, vol. xxi. p. 66. 

t Thomson and Tait, ' Good Words,' October 1862. 


heat engendered by their united collisions would only cover 
the emission of heat from the sun at its present rate for 
45,589 years. The meteoric theory is supported by Prof. 
Lockyer * and Prof. A. Winchell f, and enlarged to the 
extent of solar heat being produced by the collision of 
our sun with another star by the late Dr. Croll {, an idea 
originally suggested by Sir William Herschel before his 
speculations upon the nebular theory . 

10. The late illustrious Helmholtz, in a lecture at Konigs- 
berg, Feb. 7, 1854, accepted the theory of Laplace, stipulating 
a special form of gaseous matter represented by a state of 
infinite diffusion in which the gas was affected by forces of 
mutual and central gravitation only, but was not necessarily 
in a heated state. Helmholtz determined the amount of 
heat that would be generated by this form of gaseous con- 
densation in the sun and planets of our system up to the 
present time upon his theory. He states " that if we assume 
about the 454th part of the mechanical force remains as such, 
the remainder converted into heat would be sufficient to raise 
a mass of water equal to the mass of sun and planets 28 
million degrees Centigrade." He assumes that the greater 
part of the heat was dissipated in space ages ago. He states 
that the cooling of the earth alone from a temperature of 
2000 to 200 degrees Centigrade would, according to the 
experiments of Bischof upon basalt, require 350 million 
years. To convert the same matter from a nebular state 
would take a period beyond his conjecture. He supposes 
the condensation of the sun to continue by his attraction 
causing the falling of the surface towards the centre, and 
thereby producing a continual development of heat through 
pressure which, assuming the sun to be reduced by this 

* < The Meteoritic Hypothesis/ 1890. 

t 'World Life/ 1889. 

J 'Stellar Evolution/ 1889. 

Phil. Trans. 1785, p. 213. 


constant contraction to the density of the earth, would at the 
present rate of emission take a period of about 17 million 
years *, so that the sun in relation to the period of stellar life 
is near its point of extinction. 

11. As an important factor of the effects of conservation 
of energy in the condensation of a nebula, a law of cooling of 
masses of gas was discovered by Mr. J. Homer Lane, of 
Washington, which is given in a paper " On the Theoretical 
Temperature of the Sun " f. This law is shown in the 
following manner : If a globular gaseous mass is condensed 
to one-half its primitive diameter, the central attraction upon 
any part of its mass will be increased four-fold, while the 
surface will be reduced one-fourth. Hence the pressure per 
unit of surface will be increased sixteen times. Therefore, if 
the elastic gravitating forces were in equilibrium in the 
primitive condition of the gaseous mass, the temperature 
must be doubled in order that they may still be in equilibrium 
when the diameter is reduced one-half. Under these condi- 
tions the intensity of the heat of the sun must have increased 
with its contraction from the nebular condition. This is a 
most important consideration in showing the possibility of 
the conservation of the energy of the solar system in past 
time, during the continuous emission of heat from its former 
more extensive surface. 

12. Recently the celebrated French astronomer M. Faye 
has written a learned work bringing forward much antique 
lore upon the subject J. He adopts the theory of discrete 
matter being originally dispersed in space, following the 
theory of Kant, and of its condensation upon the thermo- 
dynamic principles proposed by Mayer, by which the collisions 
of gravitating matter about the sun produce its heat and mass 
under certain conditions. He supposes that matter drifted 

* Phil. Mag. ser. 4, vol. xi. p. 505 et seg. 
t American Journal of Science, July 1870. 
I < Sur I'Origine du Monde,' 2nd ed. 1880. 


originally in cyclones, according to the whirlpool theory of 
Descartes. He objects to the purely mechanical demonstra- 
tions of Laplace, whom science generally regards as one of 
the greatest celestial mechanics since Newton, in his theory 
showing the direction of rotation the planets and satellites 
must necessarily take upon exterior condensation through 
contraction of a rotating gaseous system. It must, however, 
be noticed in this argument that M. Faye changes the 
theoretical premises from a concrete or fluid system to a 
discrete or chaotic system wherein the particles of matter are 
assumed to be originally moving in free orbits*, in which it 
is certain that the application of the arguments of Laplace 
cannot hold. It does not appear to me that there is really 
very much difference between certain of the conceptions of 
Descartes which are applicable to the subject, if these are 
stripped of their complications, and those of Laplace. Both 
these philosophers negative a chaotic state |, and recognize 
the necessity for assuming an original fluid state to account 
for the direction of the rotation of the planets being the same 
as that of the revolution in their orbits. This gaseous state, 
surrounding the solar system in which grosser particles are 
assumed to float, is defined by Descartes as " ciel liquide dont 
les parties sont extremement agites " J, also as " corps subtile 
et tres liquid ", conceptions of a gas which do not vary 
greatly from those of Clausius and Clerk-Maxwell. With 
Laplace the original solar nebula moved in all parts with 
equal angular velocity ; with Descartes its motion was 
cyclonic, which is the only possible form of motion for a 
fluid system moving with uniform angular rotation, while 
condensing or contracting under gravitation, to enable it to 
find accommodation for the momentum of its separate parts, 

* < Sur 1'Origine du Monde,' 2nd ed. 1880, p. 264. 
t ' Les Principes de la Philosophic/ p. 147. 
J Id. p. 129. 
Id. p. 184. 


as I have shown in principle in my work on the Motion of 
Fluids . 

13. One most important work that M. Faye has done in 
this theory is to show that a discrete system of matter dis- 
tributed in space, in orbital motion in each of its separate 
parts, will, upon condensation under the action of gravitation 
to form an exterior body or planet, cause this body to rotate 
in the reverse direction to that of its solar orbital motion t 
This demonstration removes a difficulty formerly experienced 
in the acceptance of the theory of Laplace since the discovery 
of the reverse direction of revolution of the satellites of 
Uranus and Neptune. It further shows the probability that 
the widely attenuated nebulous matter which may have been 
present in space as a part of our solar nebula exterior to the 
orbit of Saturn may have condensed at an early stage into 
free particles before the concrete formation of the planets 
Uranus and Neptune and their satellites. This may be 
taken as a very probable hypothesis, the possibility of which 
appears to have escaped the powerfully analytical mind of 

14. In the hypothetical element of our knowledge of the 
extent of time which may have been taken in solar-planetary 
formation, M. Faye leaves the astronomical to consider the 
geological conditions by taking the earth's superficial strati- 
fication as an index of the entire past of the solar system. 
This is unfortunate for one who has evidently not made 
geology a serious study. In this discussion past time is 
divided into Eocene, Miocene, and Pliocene periods, all of 
which the geological student regards as recent J. This 
error, in a geological sense, is discovered and corrected in a 
second edition of his important work, but in this case it still 

* ' Experimental Researches into the Properties and the Motions of 
Fluids ' (Spon, 1881), p. 224 et seq. 
t ' Sur I'Origine du Monde/ 2nd ed. 1885, p. 117. 
| < Sur I'Origine du Monde,' 1st ed. p. 254. 


compresses solar-planetary time into 20 million years. This 
period has been proposed by eminent physicists upon arbitrary 
data, but is accepted by very few practical geologists as 
sufficient. Prof. John Perry has quite recently suggested 
for consideration much more probable data for the time 
of cooling of the earth to its present state by taking it from 
a uniform temperature of 7000 Fahr., by which, upon his 
calculation, the time would be extended to 100 million 
years *. 

Geological time, if the evidences taken from fossil remains, 
the stratification of miles in thickness by slow deposition of 
rocks, the removal of these rocks by erosion many times, with 
erasure of faulting due to plutonic action, are fairly con- 
sidered from observation and taken altogether, appears to 
carry the limit of time beyond possible conception. No short 
period of a few million years will satisfy the evidences of the 
changes occurring in the evolution of animal life alone the 
changes from one stratum to another presenting gaps evi- 
dently of much longer period than the period of deposition. 
If we take the earliest life we know of, the animal is still a 
perfect highly organized structure, which for possible evolu- 
tion indicates a long period lost to observation in fossil 
remains. No one can estimate the evidences of geological 
time unless he works in the field of geology. Let him follow 
in the excursion of such excellent societies as that of the 
Geological Association of London, stand in front of a moun- 
tain of an early Silurian period, formed in part entirely 
of comminuted shells of mollusks, with here and there a 
perfect weathered specimen. He begins to feel the immensity 
of time the generation of this life-refuse must have taken, 
although he knows his observation extends over but a small 
fraction of a long series of periods. The limit of infinite time 
is taken by some modern philosophers in the same spirit of 

* ' Nature,' Jan. 3, 1895, p. 224. 


doubt as the limit of space was formerly taken by the 

15. No attempt will be made in these pages to discuss the 
hypothesis of the diffusion of matter into space by violent 
explosions from the sun and the earth and from other cosmic 
bodies, as probably this hypothesis will have but a limited 
time popularity. The difficulty to the serious physicist, as 
shown by Lord Kelvin, is to comprehend the wonderful con- 
servation of energy we find in the sun and stars for the dis- 
pensation of heat and light. It is therefore physically beyond 
comprehension that there could remain in cosmic bodies the 
large amount of energy sufficient for the dispensation of heat 
and light, and at the same time an equal or greater amount of 
energy beyond that which is in any way evident, for the 
diffusion of matter such as satellites, comets, and meteorites 
into space by projection from their orbit foci, even if this 
would really account for their present motion and condition. 
Otherwise the projectile hypothesis is in every way in direct 
opposition to the nebular hypothesis which it will be my 
object to consider. We cannot possibly form a theory in 
which we derive energy from condensation, dispense it equally 
by diffusion and still have it largely conserved, as we know 
it to be actually in solar and stellar systems. 

16. The theory herein treated will be directed to show the 
possibilities of the concentration of sufficient solar energy, 
and of sufficient geological time in the past, to satisfy the 
direct inferences of observation. This may be possibly best 
secured by assuming our original solar nebula to be repre- 
sented by such actual nebulae as we may observe with the 
telescope. The nebulae selected will be assumed to go through 
certain changes upon condensation, the state of which may 
also be represented by other visible nebulae. In this study 
we may follow closely in the theory of Laplace with deve- 
lopment as far as possible by calculation of the actual motions 
of some part of the solar-planetary system. 


The proposed data, founded by inferences drawn from 
astronomical observation, will entirely fall in with the possi- 
bility of heat and motive energy being due to concentration 
of gaseous matter in a state of infinite diffusion in revolution 
in the solar system, as proposed in the contraction theory of 
Helmholtz. At the same time we are bound to consider 
the effects of gravity acting upon discrete matter in cosmic 
formations, found in the theories of Kant and Faye, although 
such matter may have been originally formed by condensa- 
tion of gaseous matter. The evidence of discrete cosmic 
matter rests upon the observation of the fall of meteorites 
possessed of planetary velocities to the earth, which is still 
taking place. Therefore, upon these premises there is the 
probability that throughout the formation of our solar-plane- 
tary system there were both gaseous and discrete condensa- 
tions upon our sun and planets. This may be particularly 
evident in periods, in an early discrete system of condensation 
of the very attenuated external nebula before its planetary 
condensation, and in the possibility also of discrete condensa- 
tions occurring at a period when the solar nebula by radiation 
of its initial heat fell below a temperature sufficient to support 
the gaseous state outside the present sun. These principles 
will be developed in the following pages, and the mode of 
special conditions of early condensation be discussed for con- 
temporary astronomical and for geological conditions in the 
past. Some suggestions will be carried forward, so far as the 
conditions remain active, to the present, and very hypothe- 
tically only for future periods in relation to the sun and the 



17. Time and Space. The infinities of time and space are 
not definable by any mental concept. The mind can only 
grasp the idea of a distant period of time and of a limited 
space. If we could imagine for infinite space any perfectly 
enclosed isolated space, to be filled with incandescent nebulous 
or gaseous matter, such a space would continue in the same 
state for infinite time, as there could be no loss of heat there- 
from to condense the gas by exterior radiation to form 
concrete liquid or solid matter, or to concentrate locally a part 
of the energy of the system in any manner whatever. If we 
please to imagine that matter originally existed in discrete 
particles equally distributed in infinite space under the like 
conditions, no centralized gravitation system could be formed. 
Under these conditions it becomes evident, if we adopt either 
the principles of the hypothesis of Laplace or that of Kant, 
that we require a separate volume or an original local con- 
densation to be subject to gravitational influences to form a 
star or system of stars, which volume must be isolated within 
the vacuous space, or in the surrounding matter or ether that 
must be relatively of lower specific density. 

18. Proposed Gaseous State. It has always been con- 
sidered as the groundwork of any cosmology to arrive at a 
clear definition of the original state of matter. If we rely 
upon actual observation for our data, we have evidence of 



incandescent gaseous matter isolated in space in our observ- 
able nebulae, whereas we can have no evidence of a discrete 
state of matter widely distributed in scattered small units of 
a few grains in weight, a mile or more apart, as some 
philosophers have suggested, as such a discrete system would 
be quite impossible of visual recognition. At the same time it 
is the probable condition that any isolated system of attenuated 
gaseous matter free to radiate heat into space will condense 
at a certain stage of temperature and form solid matter, par- 
ticularly if the gas is a heated form of ordinary solid matter. 
Therefore, taking an original nebula to have been in a gaseous 
state would not, in some cases, materially change the final 
results from that of a discrete system as regards the probable 
condition of condensation which may be instituted to form 
the present stars, sun, or planets, if other conditions support 
this theory. 

19. As regards the original state of matter from which we 
may conceive cosmic bodies were formed, it appears to be 
most rational for the mind to assume that matter existed 
originally in a pure or elementary state, and that it after- 
wards became mixed or combined by the action of interior 
and exterior forces acting upon it under principles which we 
generally term the laws of Nature. This purity of state for 
the units of attenuated matter may possibly be best conceived 
by taking it to be originally gaseous, as all matter can be 
shown experimentally to exist for indefinite time permanent 
in this very attenuated condition where its heat is conserved 
from radiation. Whereas, any system of discrete particles 
or dust in the presence of gravitation acting upon it cannot 
be kept in an attenuated or separate state without impression 
of an exterior force, rotative or other, to act constantly upon 
every particle as a means of separation. It becomes, there- 
fore, convenient in this discussion, without regard to theory, 
to presume some form of a gaseous state to be the original 
condition, as experiment shows that all material bodies with 


which we are acquainted may reasonably have been derived 
therefrom by reduction of temperature alone, whatever state 
or mass the final condensation may assume. 

20. Separation of Systems of Original Matter. To support 
the nebular theory of Sir William Herschel and Laplace, 
which assumes that stars and solar-planetary systems were 
formed from an extensive gaseous nebula widely distributed, 
such as we have evidence of actually in local systems of 
matter dispersed in the universe available to vision in the 
telescope, and to analysis by means of the spectroscope, it is 
necessary, as stated above, that we should assume a distant 
period of time when a certain volume or separate volumes of 
highly attenuated nebulous matter existed detached from 
surrounding space yet moving within it. The volume of 
such matter may be as extensive as we may please to imagine 
it without changing the general conditions. It may embrace 
the whole of that part of the universe we define as the Milky 
Way, or for particular evidences in detail be restricted to 
our own solar system. To support this theory it is only 
necessary that the nebula in the system that we separately 
define should have a surface boundary from which it can 
radiate the heat into space which maintained it originally in 
the nebulous or gaseous state. We have, by the effect of the 
radiation of heat from such a system of matter, the assurance of 
its constant contraction in volume, tending, by the approaching 
nearness or contiguous cohesion of its parts, to the closer 
accumulation of matter in a local focus or in local foci, so that 
matter is thereby brought more forcibly under the centralizing 
action of gravitation. The effects of the condensations upon 
such foci from a gaseous system render available, upon che- 
mical change of state or upon thermodynamic action, a store 
of energy which is sufficient in the extent of nebulae here 
considered for the formation of incandescent stars, and, if any 
limited volume of nebula be taken to be locally in rotation, 
for the production of a planetary system that may be finally 


formed therefrom, after a certain amount of radiation of its 
initial heat into space. 

21. When we conceive that our solar nebula may have 
formed a part of a general system of nebulae as defined above, 
and therefore that it may have resembled other astronomical 
nebulae, we may conclude that it was of that form, among the 
many known forms, best adapted to produce our present 
solar-planetary system upon condensation. Further, it is not 
certain that the nebula of our system, taken as a motive 
system, may not have gone through certain changes by which 
it might at various periods be represented by various forms 
of visible nebulae. Thus an originally diffused system could 
not have a nucleus or central condensation until this was 
formed, and when formed the nebula would possess a new 
external appearance. There are known nebulae which present 
irregularities of form inconsistent with condensed gravity 
systems, appearing as irregular streaks and masses of incan- 
descent hydrogen and helium. Such systems we cannot 
assume to have arrived at their final concentrated forms, 
although it is at the same time probable that they are more 
complete as gravitative systems than they appear. Probably 
the marked irregularity to telescopic vision in some of these 
masses, particularly of the spiral nebulae, depends upon the 
incandescent hydrogen and helium being the visible part of 
the nebulous mass, whereas other gaseous matter, invisible 
in an incandescent gaseous state, makes up the entire mass. 
Such matter for instance as hydrogen united with oxygen 
in the proportion to form water would be invisible under the 
conditions which would render the hydrogen visible. This 
residual matter, which the spectroscope does not grasp, may 
possibly be detected at a future time by some new method 
of analysis. Possibly it may be inferred in some cases by 
refraction and by clouding effects upon more distant objects. 
In nebulae that are so extensive as to suggest nearness to 
us, refraction may be suggested as evidence of transparent 


matter. In part of the nebula near 52 Cygni I$V15, the 
visible nebula appears as a curved streak of incandescent 
hydrogen, shown in one of Dr. Isaac Roberta's beautiful 
photographs, wherein upon the hollow side of the curve stars 
appear to be larger and much more numerous than on the 
convex side, as though the complete nebula were of spheroidal 
or lenticular form upon the concave side, invisible itself 
through its transparency, yet possessing sufficient refractive 
power to act as the object-glass of a telescope to magnify 
and bring out stars in the background which would otherwise 
be invisible in our telescopes. Dr. Roberts states that " this 
gigantic nebula is of an irregular oval character ; " and that 
" the bright side of this nebula seems to form a sharply- 
defined boundary between the stream of the Milky Way stars 
and those on its preceding side/' The photograph* for 
Plate II. e e' was taken by permission from a print, and does 
not do justice to Dr. Roberts's original negative. The hollow 
side where magnification occurs is shown towards e'. The 
dimness surrounding nebular fields being evidence of the 
presence of nearly invisible matter was pointed out by Sir 
Wm. Herschel. Possibly also such matter may surround 
and float up the hydrogen chromosphere of the sun. 

22. Tenuity of Original Matter. This is necessarily in- 
sisted upon in any system of cosmology. If we take a 
globular volume of gas extending to the orbit of Neptune 
only as the limit of our solar system and extend the mass of 
our sun and planets to this volume, we find by calculation 
that the mean density of such matter would be equal to 
about 1/166800000 that of air at the earth's surface. If 
we extend this to the mean distance between our sun and a 
near star, we should have to add many decimal places to our 
denominator. Further, it is difficult to conceive that an 
isolated system of matter as here proposed could remain of 

* ' Selection of Photographs of Stars, Star-clusters, and Nebulae/ p. 115. 


equal density throughout, unless the central heat was enor- 
mously greater than that of the outer parts. Therefore the 
gaseous matter must decrease in density in some form of 
geometrical ratio from the centre to the exterior surface, and 
this must produce a tenuity about the limits much greater 
than that shown even by the mean density of the system 
suggested above. 

The speculations of Sir Win. Herschel, as well as those of 
Wright, infer that the entire system of the Milky Way 
formed at one period a unit system of matter. This appears 
to be probable to modern science from evidences of the unity 
of chemical constitution of the stars shown by the spectro- 
scope. To account for sufficient tenuity in original cosmic- 
matter, it was suggested by Descartes that a small mass 
divided into detached particles as nearly in contact as matter 
can approach may fill a volume however large. Infinite 
divisibility of matter is, however, inconsistent with chemical 
phenomena, which are better explained by the atomic theory. 
23. In regard to the atomic theory, if we may take jointly 
the calculations of Cauchy from the motion of light in solids 
and liquids, of Lord Kelvin from certain electrical pheno- 
mena, and of Clausius and Clerk -Maxwell from gaseous 
phenomena, the mean size of the ultimate atom is about 
one 500-millionth of an inch *. In the amount of diffusion 
discussed in the previous paragraphs for space this would 
leave less than a single atom to the cubic metre. 

Therefore, if the above-stated gaseous theory is approxi- 
mately correct, it is difficult to suggest that such an atomic 
system formed our original nebula, even if the nebula w r ere 
sufficiently heated to produce a system of general gaseous 
diffusion for such atomic separation. The principles of 
diffusion may be materially strengthened if we can find 
it accordant with the inferences of science that matter 

* Lord Kelvin, Proc. Roy. Inst. vol. x. pt. 2, p. 213. 


may exist in much finer division than in the atomic state 
as theoretically defined above- 

Sir Benjamin Brodie in a lecture on Ideal Chemistry, 
delivered before the Chemical Society in 1867, suggests a 
former wider division of matter than that of the atomic state 
in a manner very applicable to our subject. He says : " We 
may conceive that in remote time and in remote space, there 
did exist formerly, or possibly do exist now, certain simpler 
forms of matter than we find on the surface of the globe 
a "> Xi ?? v> > an ^ s on. ... We may consider that in remote ages 
the temperature of matter was much higher than it is now, 
and that these other things existed then in the state of 
perfect gases separate existences uncombined .... We 
may then conceive that the temperature began to fall, 
and these things to combine with one another and to enter 
into new forms of existence, appropriate to the circumstances 

in which they were placed We may further consider 

that as the temperature went on falling, certain forms of 
matter became more permanent and more stable, to the ex- 
clusion of other forms. . . . We may conceive of this process 
the lowering of the temperature going on, so that these 
substances, when once formed, could never be decomposed 
in fact, that the resolution of these bodies into their com- 
ponent elements could never occur again. You would then 
have something of our present state of things " *. 

24. Suggestions for the Constitution of the Nebulce. 
Pneuma. The word " nebulae " which is used in Astronomy 
to define luminous cloud-like matter, particularly incan- 
descent hydrogen, helium, and carbon, will, upon the 
suggestion given above, scarcely embrace the entire early 
attenuated form of matter that the full consideration of the 
nebular theory demands, as such early nebulae must have 
been formed of the constituent parts of all chemical elements, 

* Quoted from William Crookes ; F.R.S., Address Chemical Section 
Brit. Assoc. 1886, Reports, p. 559. 


not only hydrogen, helium, or other gases that become visible 
in an incandescent state under electrical excitation. I pro- 
pose the word pneuma to specially define this most attenuated 
form of gaseous matter which may have pervaded space, 
composed of any or all the chemical elements, and which 
represented the state of infinite diffusion proposed by Helm- 
holtz. This matter would be transparent and not be visible 
in any form except when undergoing chemical combination 
or in condensation to form the visible nebula. The pneuma, 
in condensing to form the nebula, may be assumed to 
develop heat and electrical excitation, which renders the 
nebula condensed therefrom incandescent at the time. 

25. In the condensation of a pneuma system to a nebular 
one, through radiation of heat, it must be the exterior surface 
alone of the nebula where chemical action can take place, 
and where heat and electricity are developed through the 
condensation, which makes the nebula become in any degree 
visible. A further condensation of the interior of the nebula 
to form a central gravitation system or sun renders also 
this centre visible by the heat due to the pressure of the 
nebula upon condensation. The light from the centre passes 
through the transparent part of the nebula or pneuma, which 
may, or may not, at the time be undergoing chemical action. 

26. Taking the subject more in detail to meet possible 
conditions, the constitution of the pneuma which appears to 
me the most consistent with the undulatory theory of light, 
and at the same time to present evidence of sufficient tenuity 
to unite the material constitution of star systems, is that of 
perfect atomic dissociation of elementary matter to the extent 
that every line of light or shadow, as the case may be, made 
visible in the spectroscope proceeding from electrically ex- 
cited highly incandescent matter represents an active factor or, 
if we please so to term it, a distinct kind of dissociated atom 
in comparison with which the chemical atom may be con- 
sidered to represent a mass. This appears to be a most simple 


hypothesis of the original attenuated form of matter, which 
defines the factors a, %, f , v of Sir Benjamin Brodie. In this 
construction we may ascribe to each kind of pneuma atom 
in a free state or when excited by heat or electricity one 
vibrational period only. It also accounts, from the certainty 
of the great expansion in outward volume of any known 
matter to produce this dissociated state, for the possibility of 
the matter of any single star system extending originally 
to the matter or pneuma of other stars, 

27. The dissociation of atomic matter to the extent repre- 
sented by spectral lines given above was originally proposed 
by Prof. Lockyer as a possible state of highly heated matter 
made visible by special lines in the spectra of the sun and stars, 
not as representing the early condition of the nebular system 
here proposed, for which this scientist holds quite an oppo- 
site theory *. In his theory of the dissociation of matter in 
the sun and stars, Prof. Lockyer endeavours to show that the 
dissociated atoms may possibly enter into several chemical 
elements represented by many lines in their spectra, so that 
any element may lack the matter that produces certain 
spectral lines. This is proposed to be particularly shown 
in one case by the omission of certain spectral lines in the 
iron group of the sun and of stellar spectra f. That the 
same principles may be inferred of an original form of disso- 
ciation is possibly evident in a case where there is a tendency 
to a community of a certain system of material associates, 
as in the yttrium group of metals, which have been found to 
be possible of separation by the refined experiments on 
" Fractionation " by Crookes, in which he has been able to 
separate yttrium in its commercial state into five or perhaps 
eight distinct elements giving special spectral lines J. 

28. In suggesting a new word for the pre-nebular state of 

* ' The Meteoritic Hypothesis,' 1890. 
t ' Studies in Spectrum Analysis,' p, 166. 
t British Association Reports, 1886, p. 583. 


matter, it may be thought that this might be expressed by 
some term already in use, as for instance the primitive fluid 
of Lord Kelvin *, or by one of the numerous conceptions of 
the ether as that of Prof. 0. Lodge, wherein ether is con- 
ceived to be the only matter and force forming substance t, 
or by protyle, the pre-nebular matter of Crookes J. Such 
conceptions may possibly embrace the qualities of the original 
materials of the universe, but they are so entirely hypo- 
thetical that they bear no relation to experience, which takes 
its first conception from structure. The indefinitely complex 
and variable motivity of a fluid assumed to produce the 
various kinds of known matter is more difficult of conception 
than that of separate structural units. Further, if such units 
can be correlated with natural phenomena as in chemistry or 
spectroscopy, they become the legitimate groundwork for 
the elements of a theory. With this conception the idea 
herein intended to be expressed by pneuma is that it is an 
active substance composed of units which represent, sepa- 
rately or in combination, all the various properties of matter. 
These separate distinct elements, of which there are assumed 
to be a much greater number than that of our acknowledged 
chemical elements, may amount possibly to 10,000 or more 
factors or varieties. Upon this proposition it is more pro- 
bable that chemical elements may be split up into many more 
elements, than that they may hereafter be reduced in number 
by finding any more generally specialized constituent material 
or atom. 

For conciseness in the following discussion, the pneuma- 
atom, or what we may call the Lockyer-dissociation-unit, will 
be termed a pneumite. The state and associations of such 
pneumites will be now considered as the groundwork of the 
nebular theory to be proposed. 

* Enc. Brit. 9th ed. vol. iii. p. 45. 

t Lecture, London Institution, Dec. 1882. 

J Address, Brit. Assoc., Chemical Section, 1886. 


29. The pneumites at the same temperature may be all of 
equal size ; they must be very much smaller than the atom, 
probably not over 1/10,000 of its diameter. They may 
follow the conditions of Prout's or of the Newlands-Mende- 
lejeff periodic law and combine consistently with producing 
equivalents to di-hydrogen atoms in giving units of atomic 
weight. They have precisely the same capacity for heat. 
They probably possess many uniform properties which are 
common to all matter besides the special individual properties 
of each special pneumite, the active conditions of which will 
now be suggested in relation to our subject. 

30. The action of electrical excitation or of intense or pos- 
sibly original heat is assumed, as suggested by Prof. Lockyer, 
to have power to set any pneumite free from cohesion 
to other matter where this is not subject to severe pressure, 
causing repulsion between pneumite and pneumite. In this 
case a pneumite may under excitation exist as a separate 
free unit, if there is no surrounding pressure or local 
attraction acting too forcibly upon the system of which 
it forms a part to cause its combination. Upon its com- 
bination with other pneumites to enter the atomic state it 
will develop heat or electrical excitation comparable to that 
which would cause its separation. 

In the free state of the pneumite it is assumed to possess 
or attain only a single rate of vibration period for each 
special pneumite. The vibration period may depend partly 
upon the initial elasticity of the surface of the pneumite, 
which may be partly developed upon its surface as an ex- 
pansion by heat, as we know no limit to the action of this 
force, or as a repulsion under like sign of electricity, causing 
by this expansion or repulsion from contact the separation 
from other near pueumites to render it motive. This may 
also produce vibrational effects from projection through 
expansion causing the collision of one pneumite with others ; 
the elastic motivity of which may continue in its vibrational 


effects by reflection in collisions through the reaction of its 
weight or energy in attraction of gravitation or affinity 
towards its own and other matter as qualities of the special 
pneumite. Any of these conditions may distinguish a 
pneumite as a separate special factor of matter. 

31. A physical constitution of the pneumite may be sug- 
gested similar to that I originally proposed for the atom, 
that of a perfectly hard centre and a perfectly tough and 
infinitely elastic impressible and compressible coating * an 
outward condition of matter that may possibly be inferred 



from the force required to bring two convex surfaces of glass 
nearly together. The pneumites of equal size and at the 
same temperature may possess relatively different diameters 
of centres and of elastic coatings, or they may possess 
diminishing density from the centre or polarity. Thus 
fig. 2 a may represent diagrammatically a very light elastic 
pneumite of wide or slow vibrational period, c one of rapid 
period, b one of intermediate period. 

32. As regards the elastic sensitiveness of any pneumite, 
d may represent a very sensitive form in which the coating 
diminishes in density from the centre outwards ; , a pneu- 
mite with a polar axis yz, possessing vibrational influences 
unequal in different directions, f may show diagrammatically 
the expansion of the pneumite by heat, p being the limit of 
expansion by increment of temperature up to the critical 
point, p' a sudden expansion at the critical point to form a gas. 
A special pneumite may be adapted to take one form of 
electricity + or , so that it can only combine with another 

* 'Fluids,' p. 10. 


like pneumite under decrease of temperature through the 
intervention or inclusion of a pneumite of a different sign. 

33. The centre of the pneumite may be in one sense a 
universal form of gravitative matter (gravite), or this may he 
an element of it, upon which alone the amount of gravi- 
tation and cohesion depends, while still possessing other 
affinities. The resistance to combination may depend upon 
the rigidity or depth of elastic coating of the pneumite, the 
limiting extreme surface being, however, a constant of 
perfect elasticity. In the combination of two or more 
pneumites at equal temperature, #, fig. 3, may represent the 

Fig. 3. 

, . 

gaseous state at the nearest approach of two pneumites, h 
this state in combination, i two inseparable pneumites at the 
same gaseous temperature. With the like factors to those 
described above as the special characters of separate pneu- 
mites, the final compound atom may possess any observable 
qualification due to its composition. In the same manner as 
the apparently similar cell in organic life possesses the 
elements of the functions of the organ to which it belongs. 

Taken in this manner, a free atom may be considered to 
resemble in a certain degree the physical state of an organic 
being, possessing potentialities which are active only when 
it is endowed with the vital force of heat or electricity, suffi- 
cient to produce the fluid state ; but which cease when this 
influence is withdrawn or dissipated. So that the atom 
remains encysted, as it were, in a dormant state, when it 
forms part of a solid mass, outwardly sensitive only to the 
properties of cohesion and static equilibrium, so as to be 


affected by temperature and electrical excitation only to a 
limited degree. 

If there is a distinct pneumite of entirely discordant vibra- 
tional period with other pneumites, it will form a permanently 
dissociated system give a single line in the spectrum have 
no power of association or absorption with other matter, and 
be impossible of condensation from the pneuma state. The 
pneumite is assumed to be the prime mover of light vibra- 
tion ; which may be communicated through ether or otherwise 
to a distance. 

The mode of construction proposed above for the initial 
units of matter appears to me to be simpler than that 
of assuming any single factor to possess at the same time 
many distinct properties, vibrational, chemical, and other. 
Indeed, it is easier to conceive 10,000 varieties of such 
structurally simple distinct units than one endowed indi- 
vidually with the many properties of the chemical atom, 
needing Clerk-Maxwell's little demon to direct it. 

34. In the general pneuma system here proposed as being 
formed of specialized pneumites, although the pneuma may 
be invisible itself when widely dispersed in space, it may 
form in great bulk a density system by mutual attraction, 
and even possess a certain degree of absorbent or refractive 
power upon light passing through it, as before suggested for 
attenuated nebulae. It would evidently in all cases be placed 
exterior to the denser nebulae before its condensation thereto, 
which may meet Sir Wm. Herschel's suggestion that a 
nebulous appearance of stars may sometimes be caused by 
their shining through an attenuated medium "*. 

35. Chemical Action in relation to a Pneuma System. 
Upon the data just proposed, if we assume that our solar 
pneuma system originally extended to the primitive radius of 
other star systems then in condensation, we may imagine 

* Phil. Trans. 1811, p. 359. 


the synchronism of rates of vibration of certain classes of 
dissociated elements or pneumites would, by equal unity or 
multiple unity of vibrational period, promote association in 
groups to form what we recognize as the chemical atom, 
which may be compared roughly to a chord in music in relation 
to its separate notes. If the atom is complex, a better com- 
parison would be to the notes of an organ tuned to a certain 
pitch, sometimes with omission of certain notes and with 
diatonic intervals between others. The number of original 
pneumites in any chemical atom may possibly be represented 
by the number of lines in the intense spark spectrum, and 
the strength of these lines gives the relative quantity of 
special pneumites of one vibrational period in the atom. In 
this form of association it follows that any pneumite group 
or, as we should term it, chemical atom is that system of 
pneumites which can associate with the least friction among 
themselves to form more concrete matter, but which would 
eject or exclude other near pneumites moving at a discordant 
rate of vibration period. The excluded pneumites might again 
unite in simple groups of coincident vibration to form other 
distinct chemical matter. The atoms or matter formed from 
a pneuma system would remain hereafter invariable and 
initial to that particular system in which they were formed. 
If separated temporarily by heat or electrical action they 
would remain near, at positions less frictional than that cf 
entering other groups, and therefore would again unite from 
the same matter into the same atomic forms. 

36. In the above construction it will be seen that a lower 
temperature or diminished electrical excitation, by diminish- 
ing repulsion or its equivalent, will permit approach and 
constantly promote association, so that this association will 
give a mean vibrational period to two or more groups of 
pneumites of multiple accordant vibration period. Therefore 
a new vibrational position in relation to the spectrum for the 
first or earliest grouping. A still lower temperature may 


produce another less divided molecular system, leaving only 
a single group of molecules sufficiently active synchronously 
for spectroscopic observation, the vibrations sinking by loss 
of temperature or electrical excitation to a perfect molecular 
concentrated condition of restricted vibration, moving below 
a light-giving period of energy. 

37. It will be seen also from this proposition that by 
vibrational unity an element may possibly be formed minus 
certain pneumites, or with a greater or less number ; as, for 
instance, the spectrum of another star system, as suggested by 
Prof. Lockyer, may resemble our own in certain spectral lines, 
through being formed of pneumite groups or chemical atoms 
from the prevailing pneuma slightly different from the solar 
constituents. The sun or a star may possess a metal that 
may resemble iron in many particulars and yet not be in all 
chemical properties exactly like our iron, from the difference 
of pneumite composition shown in the quantity of special 
pneumites which constituted the pneuma from which it was 
formed ; but once formed it would be universal to the special 
system. Any line in a star may be omitted, or other lines 
added in an approximately like chemical element. Certain 
pneumites of accordant period may combine in several groups, 
as, for instance, certain hydrogen pneumites of the C, F, G- 
groups may be present also in nitrogen, only slightly dis- 
placed from a normal position, or even in a kind of duplicate 
motive action giving two lines instead of one, through 
collateral or rotative vibrational influences of other combined 
pneumites special to the nitrogen or the hydrogen atom. 
The sensitiveness of any special group by its collateral in- 
ternal vibrational freedom would produce sufficient vibrational 
amplitude for spectroscopic observation, whereas another 
more restricted motive group, as before stated, would fail 
in this, and therefore be invisible. 

38. Cohesion of Matter. Further, upon the above pre- 
mises, there must be a unity of vibration-period for the 


concrete chemical atom derived from the mean momentum of 
the associated periods of the pneumites of which it was formed, 
and the possibility of a like unity of atom and atom to form 
the molecule. Therefore the possibility of a unity of space- 
motion between like chemical atom and atom and molecule 
and molecule, and thereby a possibility of near approach 
when in like phase causing such atoms or molecules to inter- 
lock as it were and form denser matter, or to form a local 
cohesive system, to which surrounding free atoms or mole- 
cules would be drawn and adhere by central affinity or 
gravitational influences to form mass or visible material. In 
this manner in the nebular system, as the denser masses of 
associated matter, atoms, or molecules approach, through the 
influence of what we may term internal cohesion to gravita- 
tion-centres of attraction, the sum of these motive systems of 
discordant vibrational period, or those constituted of lighter 
vibrational momentum or of higher elasticity through sim- 
plicity of pneumite-composition, forming altogether what we 
recognize as lighter or more repulsive matter, would be 
displaced to the exterior parts from gravitation- centres, as 
hydrogen and helium are about the stars and visible nebular 
systems. The motive energy lost by the concrete association 
of pneumites to atoms, and atoms to masses may be trans- 
ferred to surrounding space. This energy, being dissipated 
through the outer attenuated pneuma, or the ether, will 
appear as tangible heat or visible light, in its encounter with 
any exterior material body *. 

39. It is possible that an atomic system may be formed of 
separate pneumites not of absolutely concordant vibrational 
period. In this case the atoms formed of pneumites of 
perfectly concordant period will be stable, and in cohesion 
also stable, as gold. Atoms formed of pneumites partly of 
slightly discordant period would be unstable, and the same in 
cohesion. Atoms formed within the greatest possible limits 

* Appendix A. 



of discordant period would be explosive. Atoms in chemical 
combination with other atoms of slightly discordant period 
would be also explosive, although they might be per se as 
regards pneumite composition of concordant period. On the 
other hand, atoms may have concordant period, although 
their pneumite composition may be partially discordant with 
other atoms. Matter formed of such concordant atoms with 
other atoms would have a tendency to associate in chemical 
composition or alloy, as nickel and cobalt, the two groups of 
platinum metals, yttrium and didymium, &c. 

40. In a vast pneuma system supposed to be in a state of 
elastic agitation through heat influences and possibly elec- 
trical excitation, which may afterwards form a solar system, 
we may assume that the primitive pneumites move in mul- 
tiple vibrational periods, #, 3#, 25#, . . . . r, or y, 5y, 9y, 
. . . . r. Since x or y may represent any number of vibra- 
tions in unit of time, these pneumites would be ready to 
unite into atomic systems. Therefore we may assume these 
concordant atoms would form the bulk of cosmic materials. 
For instance, iron with its great number of spectral lines 
may be the predominant material. This is consistent with 
its predominance in the meteoric matter which falls to the 
earth, condensed from the universal pneuma. The last 
pneumites to unite would be those at the limits of the possible 
approximately vibrational period that would be capable of 
forming atomic systems. These would bear some relation to 
the tempered notes c J, d\) in pianoforte tuning in relation to 
/, g, a. Or as ax, $x, <yx, 3a#, 5/3#, 27y# . . . r, where a, /3, y 
represent the beat periods only of the vibrational time of x. 
Matter formed of such atoms would be possibly rare in a 
cosmic system, as the greater number of pneumites would be 
selected to unite in the permanent atom-groups of simple 
multiple period. Such complex atoms in /8 7 periods would 
be possibly open to dissociation or to form separately material 
bodies, a, 3, 11 a, . . . r, or /3, 5/3, 6/3, . . . r. Under such 


refined systems of analysis as that of fractionation by Crookes, 
therefore such factors of matter would represent the meta- 
elements of that philosopher *. 

41. Observation of Astronomical Nebula. The theory of 
the nebular condition derived from observation of the bluish 
or greenish unresolvable nebulae, having regard to the facts 
revealed by experimental science, appears to be best explained 
by assuming these nebulae to be purely gaseous. The mode 
of condensation which I would suggest, which renders these 
masses visible to their extreme outline, is purely chemical in 
the form herein proposed and takes place within the exterior 
surface of the nebula only. The nebula may or may not 
possess a central incandescent nucleus. The chemical action 
results in the degradation of the pneuma to a gas or a nebula, 
under which action free electricity is developed. 

42. In the early purely pneuma state all the elements 
may combine by mixture, but still remain separate distinct 
pneumites. The condensation of pneuma to the nebular state 
causes matter to fall towards the centre through cohesion, 
where a secondary central system of illumination through 
heat may be formed by gravity, leaving an attenuated atmo- 
sphere of the lighter, more permanent gases in exterior 
position, as before stated. The effect of the superficial 
condensation is to develop electricity, just as the condensa- 
tion of water- vapour in clouds develops it in our atmosphere. 
Therefore, if we could see a nebula quite near, it would 
within its surface be continually sparkling or possibly be 
suffused with dispersed flashes of lightning. These sparkles 
or flashes would not occur in the extreme limiting surface, 
which would possibly be of hydrogen and helium in a high 
state of diffusion, but in a somewhat lower stratum, where 
possibly air or nitrogen would form a denser layer, probably 
united with aqueous vapour as in our atmosphere. These 
flashes are assumed to produce the bright lines of nitrogen 

* See Win. Crookes's important Address to the Chemical Society, 1888. 



or helium occasionally seen in the spectroscope. The ex- 
terior hydrogen and helium, in greater tension than that of a 
Geissler's tube, becomes electrically excited and forms also 
an insulator to the internal matter which is under intense 
chemical action, or under vivid electrization through the 
chemical action ; so that we may consider an apparent nebula 
as the seat of an auroral display. This superficial action of 
forces, producing light-vibrations, would detract little energy 
from a large volume of matter in a nebular system, whereas 
the radiation of intense heat from the entire system sufficient 
to produce the light we observe would detract much, and 
still not account for the spectroscopic phenomena. 

43. Condition of the Sun. As the sun is evidently at a 
temperature above that of dissociation of the chemical 
elements, it must be represented by a permanent gas, the 
gravitation of the mass causing great pressure in its central 
parts. Under such conditions there would still be the 
tendency for the exterior dissociated pneumites to unite into 
equal or multiple vibrational systems, although the elasticity 
or repulsive action of the surface of the pneumite could never 
at its temperature be overcome by pressure so as to allow of 
a very dense system being formed. Probably the constant 
tendency of the pneuma to unite into a density system in 
vibrational unity is the cause of intense chemical action, 
development of heat, and electricity. Through the internal 
friction of the heat vibrations of the system a liquid density- 
system may never be approached under present conditions. 
The temperature of the sun at present, no doubt, dissociates 
all elements, and projects the dissociates {pneumites} from its 
surface as gas where the pressure is reduced, but such pneu- 
mites cannot maintain a permanent state under the open 
radiation of the sun's surface. They therefore condense to 
nebulae and obstruct his direct rays, still moving at their 
own vibrational period, and become light-absorbing elements. 
It is probable that very refractory matter projected locally 


outwards from the sun's surface may not reach the surface 
of the photosphere, but that it condenses to nebula through 
radiation at a lower depth, producing the deeper surface 
of the sun-spots. These spots may, in this case, be possibly 
found to be formed of a close series of absorption-lines due 
to very refractory matter, which may be observable if the 
central light is sufficient to pass through them to show 
the inter-vibrational intervals. Therefore it is possible that 
the more refractory matters invisible in the solar spectrum 
of the photosphere may hereafter, with refined analysis, be 
found in the spots. 

44. In assuming the sun to have been condensed from an 
extensive nebula which extended originally much beyond the 
orbit of Neptune, it would appear at first thought that the 
lighter, less refractory matter in the gaseous state would 
remain at an exterior position. This has reasonably been 
proposed to account for the smaller densities of the superior 
planets. Upon this argument the presence of hydrogen near 
the surface of the present sun appears to be anomalous. 
There are, however, two reasons why it should appear about 
the sun : (1) The sun derived its matter from all directions. 
Tangential motion assisted in prevention of approach of 
lighter matter in the planetary plane where matter was 
placed most perfectly in motive equilibrium with gravitation. 
Whereas above the solar poles, where all matter must 
approach without tangential resistance, hydrogen would also 
ultimately approach when all the more refractory matter had 
condensed. (2) Within the pneuma system proposed the 
pneumites of hydrogen possibly possess perfect vibrational 
concord with those of a dense metal, say, palladium, and 
therefore condense to vapour with it, but at the intense heat 
upon the present sun-surface the hydrogen is constantly 
excluded, although it remains permanently about the sun's 
surface by its gravitation. 



45. Formation of Nebulce and Stellar Systems from the 
original Pneuma System in tlie Milky Way. Under the con- 
ditions suggested in the last Chapter we may assume that 
the whole system of the Milky Way formed an immense 
pneuma moving in slow rotation, the volume of which 
included the original places of the matter which surrounded 
and formed all the stars of the system. Such a system, of 
the volume the premises infer, could not be maintained static 
against gravity unless its pneuma was in a highly heated or 
motive state. In the interior of the system heat would be 
best conserved from radiation, therefore the exterior parts of 
the system alone could radiate heat freely into space to con- 
dense this pneuma into the nebular state. Taken in this 
way, the parts of the entire pneuma upon which separate 
nebular systems could condense by radiation into nebular 
star systems would be at first only relatively near the exterior 
limits of the system. The outer condensing parts of the 
system would, on account of distance, be held only lightly to 
the gravitation centre of inertia of the entire mass, and there 
would be present certain elements of tangential action from 
the rotation. The nebula would therefore experience effec- 
tive resistance ill falling towards the centre from this cause 
and from the motive elasticity of the interior highly heated 
parts, which would become by condensation denser than the 
exterior parts. Under these conditions the early formed 


nebula must either float as it were within the surface of the 
more central highly heated pneuma or be condensed into its 
mass. In either of these cases the exterior surface of the 
system becomes at the time of its condensation superficially a 
denser system than before. Eadiation of heat into space 
being constant, the surface condensation would remain con- 
stant, and as the interior elasticity would be maintained by 
the central heat, there must therefore be necessarily a certain 
amount of tensile strain upon the exterior matter under 
condensation produced by the decrease of its volume by 
shrinkage from radiation. Thus there would be a tendency 
for the exterior condensing matter of the pneuma system to 
separate into detached parts or nebular systems formed by the 
condensation of limited volumes of exterior pneuma contract- 
ing upon themselves. These condensations may be conceived 
to separate the outer pneuma system by tension, something 
after the manner that basaltic columns in formation are sepa- 
rated upon an equal gravitation plane from their uniform 
matrix. After separation such unit nebular masses would 
gradually draw their matter together into higher density 
systems, and finally by gravitation into globular volumes, the 
matter gradually further increasing in density in time towards 
their own centres. 

46. The condensing systems described above, if seen from 
a great distance, would at an early stage resemble systems of 
nebulous stars spread out upon a thinner nebular base. As 
soon as the exterior parts of the nebular system formed a 
more condensed or separate concentric nebular star system, 
these nebulous stars would further contract upon themselves, 
and then the interspaces thereby produced would permit the 
free radiation of heat from the more interior parts of the 
entire system of the pneuma, so that upon the same con- 
ditions another concentric stratum of condensations more in- 
terior than the first would again form a nebular star system, 
and this again another in like manner still more interior, 


until the whole system of pneuma here considered was con- 
densing through radiation of heat to nebulae and finally into 
stellar systems under initial gravitation with intense che- 
mical action. Upon this principle the pneuma would be 
separated into unit nebular star systems, and finally into 
individual stars each radiating heat proportional to its con- 
densation into universal space, as it does at the present time 
in our Milky Way. 

Certain local systems as here defined possessed of great 
central density, if seen from remote distance, would resemble 
in the whole or in partly advanced stages of condensation 
some of the globular or lenticular clusters such as Messier 3, 
13, 15, 53, 92, and others. 

47. Suggested Motive conditions of the Original Pneuma of 
the Milky Way. If this pneuma was originally motive by 
rotation about a centre of inertia or otherwise, as just pro- 
posed, its separate units, detached by local condensation, 
would be also motive by continuity of the original momentum, 
which in a perfectly fluid system would be diverted in any 
direction that at the time presented the least resistance to its 
motivity. There is not, however, in the Milky Way the 
perfect symmetry necessary to infer an original uniform 
rotative pneuma formed under simple condensation. Our 
Milky Way is apparently an immense flattish bifurcating 
plane of great depth formed of stars unequally distributed. 
Another mode of formation may be suggested which, although 
entirely hypothetical, or may be fanciful, is not altogether 
inconsistent with the wonderful variations in the forms which 
exist at the present time in visible nebulae. These proposed 
conditions may also have partly induced the separation of 
the pneuma system into unit star systems, and are quite 
consistent with the motion of matter upon the nebular hypo- 
thesis of Descartes, which is so ably supported by M. Faye. 

48. Assuming that there were vast pneuma systems moving 
originally freely in space, the tendency of such separate 


systems in isolation under the influence of gravitation would 
be to form spheroidal systems. The flattish form of our 
Milky Way may possibly indicate that two such spheroidal 
systems of immense volume at an early period may have 
drifted together. Such a collision as would be produced at 
the meeting-surface would form at once a relatively superior 
density plane where the matter of the two systems would be 
united, and henceforth the centre of gravity of the two 
systems would be changed to the centre of meeting-plane of 
the two systems. The pneuma systems being in the highest 
degree elastic, would continue the momentum of their original 
projection of mass in direction normal to the plane of contact 
for many millions of years after impact, continuing also 
to compress and flatten out the matter of the original 
meeting-plane. The pressure in this plane would be increased 
independently of original momentum by the action of the 
mutual gravitation of the two systems in approaching to unity 
throughout the meeting-surface. One effect of the compres- 
sion at the meeting-surface described above would be the 
development of intense heat and electrification. The elec- 
trical excitation would diffuse itself through the entire mass, 
possibly rendering it luminous to its extreme limits. 

49. The result of such a collision after n millions of years 
would be the formation of a vast lenticular pneuma more 
intensely heated in its central plane, which would be more 
dense from compression and gravitation than the surrounding 
parts. If the two parts did not entirely combine, a bifurcating 
system might be formed. The whole system would under 
any condition possess a momentum compounded of the original 
momenta of the two pneuma systems from which it was 
formed, but in its separate parts it would possess motivities 
consistent with the conditions brought about by the collision 
and reflective reactions of parts of the perfectly fluid matter 
of the pneuma systems. Without such a collision as herein 
depicted it is possible that the condensation of the stellar 


pneuma systems of the parts of the Milky Way would have 
been much slower than they were, so as possibly not to have 
advanced at the present time beyond the condition of what 
may be the present state of outer matter about the galactic 

50. Cyclone-inducing conditions. Assuming the pneuma to 
be a most perfect fluid and elastic system of matter, upon 
the meeting of two volumes of such matter, indepen- 
dently of any initial rotation it might possess, must have 
moved under pressure at the meeting-plane in every or any 
direction which at the time offered the least resistance to the 
continuity of its initial momentum. Therefore, in the meet- 
ing of two fluid systems as proposed above, the direction for 
continuity of motion of the outflow of these systems from the 
contact plane, where the pressure would be greatest, being 
supported by the momentum of the following parts, would 
drive the compressed motive pneuma into directions normal 
to the pressure. Under the conditions given the outflow 
would be along the plane of contact and outwards in every 
direction from the centre of greatest compression, thus pro- 
jecting the fluid into a current plane. Such a motive plane 
or current, as I have shown by many experiments in my work 
on ( The Motion of Fluids ' *, engenders motive whirls in the 
contiguous parts of the surrounding fluid, this form offering 
the least frictional resistance to continuity of motion of a 
fluid moving within a like fluid. These whirls, if formed on 
the principles suggested, would in their turn engender other 
friction-whirls exterior to themselves, until the entire system 
became one of complete eddying motion, as it is termed. 

51. As this form of fluid motion, originally investigated by 
myself, is not very generally known, it may be as well to 
demonstrate this hypothesis by giving an illustration which I 
take from an experiment in my work on the Motion of Fluids, 

* ' Fluids,' pp. 227 and 310. 


in which the resistance to a flowing fluid in a plane of pres- 
sure is applicable to the conditions suggested above. Although 
the experiment referred to is on a very small scale, I have 
shown that motive systems of fluid are proportional and 
irrespective of scale, so that the same principles extend to 
the surface motions of the North Atlantic Ocean as hold good 
in the small experiment I describe * : " Take two plates of 
perfectly clean glass, say about six inches square, and firmly 
cement a border of card about y^ of an inch in thickness 
round three sides of one of the plates with marine glue. 
Then cement the second plate to the first. By this means 
a very thin waterproof trough will be formed, open at the 
top. If we now fill this with clean water and place it 
upright in a groove in a piece of stout wood, the experimental 
apparatus will be complete. 

tl To observe the desired effects, the trough should be placed 
before a window to transmit light through it, when the 
following phenomena of original fluid projection may be 
observed by the projected slightly heavier liquid diffusing 

* ' Fluids/ p. 866,2. 



itself in a current, through the constant force of pressure of 
its small excess of gravity. 

" Take a pen full of ink and place this gently upon the 
surface of the water in the trough. The ink as it descends 
slowly, by excess of gravity over the water, will he found to 
divide constantly upon the resistance which opposes its direct 
projection, the divided parts having insufficient momentum 
of projection to move the lateral fluid tangentially to pro- 
duce extensive lateral whirls. After the first division of the 
descending fluid the divided parts again divide, and these 
again, so that by this constant division and after subdivision 
the projected fluid takes a tree-like form, the terminals 
being in spiral rotation. The illustration (fig. 4) was taken 
by transmitted light, exactly following the outlines of the 
projection of ink in the thin trough described ; it is therefore 
represented of the observed size " *. In this experiment the 
sides of the trough are assumed to represent the pressure 

Fig. 5. 

52. The principle of lateral diffusion of whirls is seen 
contiguous to any flowing stream in lateral water : the sketch 
(fig. 5) was taken from the flow of water through an arch 
of London Bridge. 

* < Fluids/ p. 314 ; id. p. 310. 


In this matter we might consider the separate motive 
cyclonic volumes of nebulae in a very attenuated system to be 
an early form of projected matter inducing rotation in stellar 
formations, as in the theory propounded by Descartes and so 
ably supported by M. Faye. These motions must, however, 
be considered to take place strictly within the pneuma, and 
would not be the cause of its condensation or separation into 
stellar systems, the conditions of which as effects of radiation 
have been already discussed, 45. 

53. Formation of Spiral Nebula and Stellar Systems. 
Taking any separate unit system of pneuma as a part of the 
system just proposed or any other, we will assume this to be 
in uniform angular revolution, its exterior parts moving with 
a velocity below that which would project them in an orbit. 
The outer matter of the system being assumed to condense 
through radiation, then particle would approach particle as 
the near parts of vapour in the atmosphere do to form rain- 
drops. Such particles or drops would follow in the general 
drift of the residual vapour of other matter that could con- 
dense only at a lower temperature as rain-drops fall through 
air. Under these conditions the earliest condensed outer 
matter would drift as a discrete system in showers or con- 
gregations or concatenations under the surrounding resistance 
in spiral lines towards the centre of the system with increas- 
ing velocity owing to acceleration by gravity *. Such con- 
gregations of material discrete particles under the influence 
of mutual local attractions would present a feathery or floc- 
culent appearance on the breaking up of the outer pneuma of 
the system. They would possibly be illuminated by friction 
on electrical excitation. As soon as the isolated flocculi 
drifted in spiral paths sunward or to the star-centre, other 
surrounding condensing matter would follow in the same 
direction by initial gravitative influences, and form currents 

* Newton's ' Principia,' Lib. ii. prop. xv. 


in certain positions piercing through the minor resistances of 
the lighter interior gaseous matter which condenses only at a 
lower temperature. In this manner a condensing nebular 
system moving in spiral lines sunward would appear, if seen 
from a distance, to be a perfectly stationary system, although 
its condensation might be progressing at a rapid rate, as the 
same position of spiral stream-lines of drifting flocculi would 
be constantly conserved locally. This may be the most 
general form of nebular condensation. 

54. Solar -Planetary inducing conditions in Spiral Nebulce. 
If the original tangential motion of a pneuma system, 
condensing with acceleration of its flocculi through gravita- 
tion into a spiral nebular system, attained sufficient velocity 
through gravity to maintain the flocculi, drifting as above 
proposed, at a certain position in an orbit exterior to the 
central system of the pneuma, this matter although in revolu- 
tion would remain in equilibrium at a certain distance from 
the central sun or star, just as though it floated on frictionless 
water. It would thus draw its separate parts together within 
certain limits by its own internal initial gravitation and the 
drift of following parts, so as to form a separate nebular 
condensation or zone exterior to the centre, which thereby 
ultimately would become in a condition to form a planet. 
The direction of rotation of such an orbit-zone or planet 
will be discussed hereafter. In the above we may see 
clearly the possibility of a partially discrete system of 
condensation of exterior matter becoming a factor of solar 
nebular planetary formation. 

Upon the conditions proposed in the above and the pre- 
ceding paragraphs, we may suggest that, in the condensation 
of the volume of pneuma necessary to form a solar-planetary 
system, a nebular system may be formed at a certain period 
by central condensation, which will maintain the nebular 
condition about the centre, but that this may be surrounded 
at an early stage by discrete condensations falling thereto, 


formed by local exterior condensations due to excess of 
ratiation of heat from the periphery of the system, so that 
the purely nebular condition for the entire system may appear 
in some cases in an early stage only. In this construction 
we have certain factors of the nebular theory of Herschel 
and Laplace in conjunction with the discrete theory of 
Kant and Faye, which may be possibly observed active in 
such nebulse as that of Messier 51 in Canum Venaticorum, 
Plate II. /. 

55. The Permanency of separate Star Systems after 
formation. If we take gravitation to be a force active at all 
distances of centres of masses according to a given law, its 
motive energy being in inverse proportion to the squares of 
distances, and if we conceive condensations to be originally 
nearly equally distributed within the limits of a unit system ; 
then all these aggregate or separate systems of matter, 
although originally in perfect equilibrium, would drift, 
though slowly, towards one another. 

56. We may, however, assume other conditions to prevail 
under which the permanency of separate stellar systems as 
local aggregates, when once formed, might be maintained. 
We may take gravitation to be inactive beyond a certain 
radius as a possible condition under which stars once formed 
would remain permanently fixed in space with relation to one 
another. It is, however, more cogent as the law of gravita- 
tion has been found to hold exactly under every test of 
experimental observation within the limited though extensive 
areas circumscribed by the orbits of double and multiple stars 
of long period to assume that this force follows the law of 
inverse squares for all distances. Under this condition a 
stellar system may nevertheless be permanent if we assume 
a universal revolution of stars about a common centre of 
inertia, as originally proposed by Wright and afterwards by 
Sir William Herschel, the principles of which are exemplified 
in our own planetary system. This does not, however, 


necessarily suggest a unit centre of revolution for all stars, or 
even for those which compose the Milky Way. The cyclonic 
systems just proposed would at least be inconsistent with this. 
There is space enough for many centralized gravitation 
systems within the more universal system of the Milky Way, 
in which the stars are placed at distances from one another 
far too great to render gravitation active upon the members 
of any separate system beyond producing slight variations in 
the forms of the stellar orbits surrounding the nearest or 
strongest central attraction ; although there may be even 
over and beyond this a general drift or circulation of the 
entire system or any part of it in which the separate systems 
here considered are embraced. 

57. The influence of the knowledge of our solar-planetary 
system acts so powerfully upon the mind, that it is difficult to 
conceive the action of gravitation upon cosmic bodies other- 
wise than with certain elements of tangential motion by 
which they move symmetrically in a plane about a centre of 
inertia. In the wide distribution of stars in all directions 
this may not be universal. It is quite possible and consistent 
with an original fluid state, that stars may not always form 
part of a closed system drifting in a circular or elliptic orbit. 
Such orbits may be contorted in any way by surrounding 
influences and still retain the elements of original sidereal 
momentum. The evidence of the variety of form taken by 
gravitative matter in the distribution of stars in star clusters 
and in some nebulse appears to show that stars and matter 
move sometimes in a system of stream-lines, following one 
another in a form we may term pearling (as pearls of dew) . 
This applies to the actual appearance only, not to the mode of 
formation. Sometimes these pearls or stars drift in spirals or 
cyclones, in like manner to the matter that forms the system 
of bivolutes in M. 51 Canum Venaticorum, M. 74 Piscium, 
1$ I. 168 Ursse Majoris, as shown in Dr. Roberta's beautiful 
photographs. We can but conceive that there must never- 


theless be beyond this throughout all matter, if gravitation 
be universal, a tendency to approach of mass to mass which 
can only be resisted by a motion in discrete bodies in which 
there are certain elements of tangential direction, and this 
must ultimately tend to cause the whole of any system of 
matter to approach a symmetrical form of orbital motion 
about its centre of inertia, which may include the orbital or 
other motions of its separate parts, as in the case of the sun 
and planets ; but the time necessary for such a complete 
formation in the universe would appear to be so far infinite 
that it may never nearly have approached completion in any 
large division of the stellar system. 

58. Distances under which Gravitation may be active. It 
may be difficult in some cases to imagine the disturbing 
influence of gravitation at so great a distance as we know 
matter to exist in stellar space, as inferred above. But, as it 
is proposed that the action of gravitation is unlimited in 
space, acting upon free bodies according to its law of accele- 
ration, then a small attraction upon a very distant free body, 
where the general composition of motions permits approach, 
will produce a great velocity in a period which may be 
relatively short for past time. Assume any free body, at a 
very distant period, to be moving towards our sun with a 
velocity of one centimetre a year, which we may take as a 
quite impalpable motion in space ; still in one million years, 
if the body remain in relative position towards the sun as 
regards exterior influences, it will have attained a velocity of 
approach of ten kilometres a year ; and this velocity would 
increase with time in like ratio, until in n millions of years, 
some small fraction of the past, it would fall towards the 
sun with velocity equal to or greater than the highest ever 
observed within our system. 

59. Suggested action of Gravity in time in the formation of 
Circular Orbits. If gravity is instantaneous for all distances, 
an orbit once formed must remain constant in relation to its 



focus for all time if it moves in a perfectly frictionless 
medium. But as the orbits of certain planets and satellites 
are so nearly circular, it would appear that there must be 
some reason for this in a circle-inducing quality as a property 
of the forces by which these bodies are directed. There is 
no doubt that in the condensation of a spheroidal gaseous 
system of nearly the diameter of a planetary orbit, as in the 
theory of Laplace, that circular or elliptical revolution might 
be brought about by the resistance of the gaseous central 
matter to change, owing to uniform angular velocity in all 
its parts. This will be again considered. But a second 
factor of resistance might reasonably be derived from a time 
element in the action of gravity in a manner also originally 
proposed by Laplace. In this proposition, as time is infinite, 
small constant forces active in the past may have produced 
great effects by the present time. 

60. It becomes, therefore, more rational, whatever we take 
space to be, vacuum or ether, to assume gravity acting 
through this space in time, say, with the velocity of light, or 
with much greater velocity. It is not in this case probable 
that any time element of attraction could be detected if the 
distance remained approximately constant, as we may easily 
imagine the action of gravity to be induced *, and that after 
induction it remains a constant for constant distance. The 
change of position in approach or recession may produce a 
change in the gravitation force of induction, and this may 
take time. Thus, as an instance, in the approach of a comet 
to perihelion under deferred increasing factors of gravity, it 
would arrive somewhat behind its time, but as the attraction 
would still act in like inverse ratio after it had passed peri- 
helion, it would retard and deflect the body if originally 
projected in a parabolic orbit into an elliptic orbit after this 
passage. If the body moved in an elliptic orbit, the action 

* Appendix A. 


due to a time factor in gravitation would decrease the 
eccentricity at every perihelion passage, and increase the 
time of revolution in approaching constantly to a more nearly 
circular orbit. The circle being the greatest inscribed area 
due to a given mean rate of tangential motion, therefore we 
may imagine it the state of least internal resistance and of the 
ultimate motive equilibrium for the orbit of a celestial body 
in relation to a single centre. 



61. Separate Stellar and Planetary inducing Conditions in 
the Nebulae, according to the amount of Rotation. Under the 
conditions already discussed, assuming that any isolated 
volume of pneuma may form a part of the universe, we may 
take it as probable that it will form a system of matter which 
will contract in volume by radiation directly under the influ- 
ence of gravity until it forms a central gaseous sphere or 
spheroid. This, if of sufficient size, would form a nebulous 
star in an early incandescent state of condensation. We may 
assume that such a system if it has no rotation, or if its 
rotation is slow, resulting either from the original motion 
impressed upon the nebula or from matter gravitating towards 
it from all directions with nearly equal momentum, so that 
from this reason the combined orbit and mass of the motive 
parts by the simple action of gravitation produce a globular 
system in the nebulous state, such a system would form a 
single isolated star moving or stationary in space. 

62. On the other hand, if the special nebular system con- 
sidered is in rotation or in cyclonic motion derived from this, 
matter would approach the centre with much greater facility 
in every direction other than in a plane extended from its 
equator, where the tangential impulse of surrounding matter 
would produce the greatest resistance to the centralizing 
action of gravitation. So that in this case a star might be 


formed of all surrounding matter except that about the plane 
of its equator. This system at a certain stage of condensa- 
tion would therefore appear, if viewed from a distance in a 
direction nearly normal to its poles, as a planetary nebula 
surrounded equatorially by more attenuated matter. Such a 
nebula would be in outline of oblate spheroidal form if the 
equatorial nebular matter were evenly distributed about the 
equatorial plane. (Plate II. a, Great Nebula in Andromeda ; 
6, Messier 32 ; c, $ I. 200 ; d, M. 81.) 

63. The spheroidal form of such a nebulous condensation 
might be modified to any extent by unequal distribution of 
the nebula about its focus into a spiral, lenticular, or 
discoidal shape, its parts being still held with equal perman- 
ence by their tangential impulses. Either of the above- 
described systems, if in equilibrium about its centre with its 
peripheral parts moving with sufficient tangential velocity to 
maintain an orbital position, would be ready to separate its 
peripheral matter into ring-zones or other separate motive 
systems upon further central condensation, and become 
finally in a state to form a solar system such as our own with 
planets of greater or smaller mass inter alia moving in orbits 
about its centre, the conditions for the formation of which 
will be considered presently. 

64. Limits of a Solar-planetary-cometary System. We 
assume that after separation of any complete pneuma or 
nebular system from the universal pneuma, the initial action 
of gravity within this separate system would immediately 
commence to form a central condensation of greater density, 
as before proposed, which would react upon surrounding 
matter in proportion to its mass and inversely to the square 
of its distance from any part of the surrounding widely 
distributed matter. 

Suppose an original system of pneuma in the remote past 
taking one direction only, extending from S to S', fig. 6, and 
that two equal centres of condensation, S and S', were after 


long periods slowly formed as star systems upon principles 
just discussed. The neutral position at which a particle or 
mass would be originally equally solicited by the gravitation 
of both systems would be at X. This would be a point of 
early condensation from open radiation, and at the same time 

Fig. 6. 

() . oc a." c c' c" V 6' 6 (S) 

one of perfect equilibrium in space. From the equal attrac- 
tion of S and S' upon the condensed particle, a particle or 
mass at c" would be very slightly influenced to move towards 

5, a particle at c more influenced, and at c still more so. 
Thus, assuming gravity universal, this would give attractions 
to matter to fall towards the centre S from points of rest 
intermediate between S and S' with velocities inversely 
proportional to the squares of the distances of S and S 
respectively active upon it. 

65. Like attractions would also hold with respect to the 
star system S' as regards particles at the relative distances 

6, b'j b". Therefore all matter from S to X, which we may 
take as the original radius of our own solar pneuma system, 
would ultimately form a part of the gravitation system of S 
with predominant influence over S', and all matter from S' 
to X would ultimately form a part of the gravitation system 
of S' with predominant influence over S. Therefore it would 
be impossible that any body placed between X and S' should 
be attracted towards S, or in fact that S should form the 
focus of any orbit of greater aphelion distance than X in this 
direction. It is further seen upon this hypothesis that 
cometary or other matter could only move within its orbit at 
the furthest from the position X in relation to another star S'. 
If matter does not fall from the positions c, c\ c" directly 
upon or towards S, there must be a tangential impulse upon 


such matter in relation to S which directs it into an elliptical 
orbit. Under any condition, if moved it could attain no 
greater velocity from gravitation in passing its perihelion 
about S for projection afterwards than that due to the 
velocity acquired in falling the distance to S, even in a 
perfectly frictionless medium. We may therefore note that 
it is not possible, if gravitation is active for all distances and 
matter is condensed from a uniform pneuma or nebular state, 
for any body or comet to reach our sun from space other 
than that which once formed a part of our own pneuma 

66. Action of Gravity on distant Condensations within a 
Solar System. As we may assume that the condensation of 
the exterior matter of the transparent pneuma would be more 
rapid from radiation of its original heat into space than that 
of any more centralized denser matter or gravitation-centre 
included within it, a considerable volume of outward pneuma 
might condense on local foci, as before proposed ; and these 
could reach the central space only very slowly from their 
distance under the slight influence of central gravitation 
acting upon them. So that a considerable mass might be 
forming at S, the sun-centre of the gravitation system, by the 
attractions of a, a', a" through the centralizing force of S 
before any effective movement thereto was induced in the 
distant parts c, </, c", assumed to be partly under the influ- 
ence of the star S'. Nevertheless, although the near matter 
a, a', a" would experience the centralizing influence of 
gravity sooner than the distant parts, the interior heat of 
the system may be assumed to be conserved and supported 
by that released by the condensation upon the central nebula 
as before proposed ; therefore it would remain a gas for a 
long time, producing a pressure only about the central sun. 
The entire condensation of the nebulous sun would be, never- 
theless, only in proportion to the radiation of heat from the 
exterior parts of the nebulous system. 


67. With respect to our owii solar-planetary system re- 
garded as a rotating condensation under the conditions stated 
above, we may consider the central nebula formed by pneuma 
condensation at a point when its extent was represented by 
N N', fig. 7, which may be a space including the orbit of 
Neptune, shown centrally transverse to its plane. Then, if 
we now limit the extent of directly condensing gaseous sun- 
forming matter to N N', the exterior matter represented by 
C" C' C being subject to greater radiation of initial heat and 

fig. 7. 

C" C" 

receiving less heat from the condensing sun, these bodies or 
parts C, C', C" might condense upon themselves and draw 
matter together as before suggested and form separate masses 
(meteorites), or, if moving in nearly contiguous parts in 
revolution upon a centre of inertia, form more extensive 
material systems, flocculi, or comets, the probability of which 
will be discussed hereafter. 

If the central nebulous system of 1ST N' were conservative 
of initial heat in a certain degree, as here proposed, for a 
time, the free separate particle or mass C attracted to the 
central nebula might in time enter the nebula N N' and be 
absorbed therein. If it did so, it would become heated by 
the friction caused by the resistance it would encounter suffi- 
ciently for it to become again expanded to gas and afterwards 
form a part of the central nebulous system, increasing its 
density thereby. This would not occur, however, without 
producing a further condensation about the position of entry 


of the mass and of motion within it, leading to irregularity 
of constitution of the nebular system. If the included mass 
projected into the nebula were sufficiently great, it might 
become an inducing factor of centralizing planetary or 
satellite aggregation or bring about a disturbance of motive 
directions of the matter within the system. 

68. If we take the condition of matter falling from a 
position C' further from the centre than that considered 
above, the gravitation upon this point being less active and 
the influence of the gravitation towards S' more active, this 
might not attain a movement in space sufficient to reach the 
central system until the latter had contracted in volume to a 
radius represented by m m, which we may take for demon- 
stration as the cross section of the sun's nebula when this 
extended to the orbit of Mercury. In this case, if the 
eccentricity of the orbit of the matter C' caused its perihelion 
distance to come within m m 1 ', the mass C' would, by friction 
arising from resistance, be retained in the nebula and increase 
its density, or it would move in spiral lines towards the sun- 
centre with velocity in proportion to its momentum and 
gravity to this focus into the resistance of the surrounding 
matter. If its perihelion distance were greater than the 
radius m, so as not to come under the sun's attraction suffi- 
ciently to be deflected from an elliptical orbit, the body would 
then move in this orbit constantly for all time and become a 
permanent planet or comet or meteorite of the system. 

It is seen in this that it is only matter endowed with a 
considerable tangential momentum from any cause that can 
form a permanent planet, comet, or meteorite of our system, 
and that the greater mass of condensing matter around the 
sun at an early period, particularly that which was formed 
outside the mean planetary orbit plane, would, in all proba- 
bility, fall directly or indirectly into the sun's nebula and 
become a factor of sun-formation. 

69. If we consider as an extreme case the condition of 


another mass C" still more distant from the sun, we may 
conclude that the small movement induced by gravitation, 
depending upon the difference of the separate attractions of 
S and S' (fig. 6), at this distance would not permit it to 
reach perihelion until the time when the sun had attained its 
present relatively small volume. In this case, supposing it 
possessed any tangential momentum, the probability of its 
falling into the immediate solar nebula would be relatively 
very small. Such a body would therefore move in an 
elliptical orbit and become a permanent comet or meteoroid 
of our system. A condensed mass at a still greater distance 
from the sun, and nearer S', would not in the past time have 
reached our sun, so that it may only at a very remote period 
become what we should recognize as a comet if of sufficient 
outward volume. 

70. In the above construction, in considering exterior 
matter to be separately condensed and to reach the solar 
focus in time proportional to its distance, this distant exterior 
matter would be little affected by the gravitation-figure of 
the central spheroidal nebulous system assumed to cause its 
orbit to be drawn toward the nebular plane under conditions 
originally suggested by Kant. Therefore the exterior con- 
densations of a pneuma system would be projected at an 
angle to this plane, with the reservation that the tangential 
momentum would be less proportionally to the increase of 
angle to this plane, if the original pneuma were ever in 
uniform rotation, so that the orbits of comets approaching the 
sun at angles considerably inclined to the mean solar- 
planetary plane should be more eccentric than those at 
smaller inclination to this plane, so far as the above stated 
conditions hold. 

71. Direction of approach to the Sun of matter from the 
exterior of the original Solar-planetary Nebula. Formation oj 
Orbits. Let A and B, fig. 8 3 be two stars, or suns, at the 
intersection of the lines A B, d d r be the point of equilibrium 



between these suns, where would rest in static equilibrium, 
the gravitation of A and B being equal upon it. Place any 
number of particles in line at right angles to A B from d 
to d'. Then gravitation acting upon any of these points 

Fig. 8. 

a, b, c, a', b f , c', considered as free bodies, would cause them 
to fall to 0, and beyond it if the impulse was sufficient, so 
that they would oscillate about constantly in the mean 
gravitative tangential plane of A and B. 

72. If we make A a superior attraction over B, either from 
nearness or excess of mass, then matter attracted from 
would fall in direct line towards A. Any other matter in 
one of the positions a, fr, c, a', 6', c', although it would 
receive less attraction to B than to A, would move in angular 
direction towards A in composition with the pull of B upon 
it. Therefore the composition of forces A and B would 
induce a definite amount of tangential motion upon all the 
points a, 6, c, a', Z/, c' moving towards A. As the objects 
represented by points approached A in moving in their 
orbits, the influence of B would diminish in gravitation 
proportion ; A becoming entirely dominant when the matter 
of any of the points approached the perihelion of the orbit 
whose eccentricity was induced by the tangential element B 
acting upon it. 


73. It is also seen by the above diagram that if the masses 
are distributed at equal distances upon the tangent d d' 
there will be many such masses a, b, c, a, b', c', influenced by 
the attractions of B in falling towards A, whereas in the 
direct line there w r ill be no tangential influence from B. 
Therefore, the general paths of bodies falling from space 
from surrounding attraction under a superior attraction, will 
possess elements of tangential motion in reference to their 
predominant attractions which will induce them to follow 
elliptical orbits, and the case of a body falling directly from 
distant space upon the sun or a star will be exceptional. The 
principle here shown by diagram applied to two stars or suns 
will apply equally to any number taken in separate pairs 
distributed as they may be in space. 

74. As regards the direction of the orbit of any free body 
passing under the influence of the sun's attraction, it will be 
seen that the amount of tangential motion induced by B upon 
d' acting in the direction shown by the arrow e' will approxi- 
mately direct the path of the body in the line p' p', in its 
orbit in a left to right direction at perihelion round the sun 
and back to its first position. In the same manner, a body 
projected to A from d with the tangential component induced 
by B in the direction of the arrow e, will follow the path p p 
in a left to right direction in the plane of its orbit. 

75. If we regard any initial motion of the circumscribed 
system about our sun in the direction of the arrow " as an 
original nebular condition, then we may assume that the 
influence of B upon matter projected towards A is neutralized 
at some point, say 6, so that b will fall more directly towards 
the sun A than 0, and the scale of distance for projection 
from this point b for the other points d, c, a, a' will be equal 
to that previously defined for the point as regards tangential 
action in inducing direction of revolution. In this case the 
whole system of separate attractions as motive forces may be 
assumed to be displaced by this initial direction of motion, 


and the preponderance of direction will be that of the initial 
tangential motion of the whole system, but not of every 
element of it. 

76. The above scheme, fig. 8, which is assumed to be that 
of the form of gravitation action between our sun and every 
near star, may be placed at any angle or direction to the 
polar axis of the sun where the star appears. If matter in 
the positions a, 6, c, d, a', b', c', d' , shown by the diagram, 
were projected at an early period near the mean planetary 
plane, the exterior body would fall into the planetary nebula, 
combine with it, and enter into composition with its motion. 
This must have been the condition of some of the early 
comets, assumed to be flocculi condensed within the superior 
influence of the sun's nebula ; and as these condensations 
must have formed in all directions with regard to the sun, 
the comets left at the present time in our system must be 
much fewer or more particularly of much less mass in the 
mean direction of the sun's equator and the planetary plane, 
than in the direction of the solar poles. This condition is 
not, however, absolute for all points, as a near star in the 
direction of the solar pole or any other direction would as a 
gravitation centre cause much less matter to fall directly 
towards our sun than that which would fall in the direction 
of a more distant star or from an intermediate space. This 
subject will be taken into consideration further on. 

77. In the construction shown in the diagram, fig. 8, it 
must be clearly observed that unless the entire system is in 
motion in relation to a gravitation centre, as shown by the 
arrow 0", the sum of the deflections of the points a, 6, c, 
a', b f , c' towards the plane AB would be equal ; so that no 
rotation would be imparted to the sun A or a planetary 
system connected therewith by the momentum of the united 
masses of this matter if it was impressed upon the sun's 
nebula at any period. This does not appear to be clear to 
some authors who have considered the subject without 


proposing clear definitions which, it is hoped, the diagram 
may supply. Kant suggested that the sum of collisions from 
exterior matter attracted from all directions towards the sun 
from space would direct his revolution and that of the planets 
in one direction as a resultant. Herbert Spencer adopts this 
view *. 

78. M. Faye shows that the mean momentum of all 
surrounding bodies drawn in direct line by gravitation would 
not impart any revolution whatever to a central system f. This 
is herein demonstrated ; so that although all comets may 
be considered as extreme condensations that take one or the 
other direction of revolution in long elliptical orbits, upon the 
principles discussed above, yet, if they were retained within 
a nebula surrounding the sun at perihelion, unless the nebula 
possessed original rotation, the mean motion produced upon 
the rotation of this central nebula would be approximately 
nil. Taken in another form, excluding original rotation, we 
might conclude that the probability is that the mean momentum 
of all the comets of our system is approximately ra7, there 
being possibly as many, or, more exactly, as much exterior 
matter moving from space in one direction as in the other. 

79. The Formation of a Planetary Plane. Under the 
above-stated condition, during the time that the sun remained 
an extensive nebula any exterior body entering this nebula 
at high velocity must have become dissociated and incor- 
porated therewith by the heat engendered in the nebulous 
matter through friction, its initial momentum being com- 
pounded with that of the nebula into which it was projected, 
as before stated. Therefore, all exterior projections into the 
sun's nebula will be drawn towards a line passing directly 
from one sun to another predominant sun as the linear 
direction of greatest attraction, and upon this principle the 

* Nebular Hypothesis," The Westminster Review, 1858. 
t ' Sur POrigine du Monde,' 1885, p. 134. 


final mean equatorial plane of revolution of the solar system 
must have become that of the mean directive influences of 
attraction of all the near stars or other matter that surrounded 
it, at the period of its formation, in combination with the 
momentum in revolution of its original pneuma or nebular 
system. This does not, however, infer that the sun itself 
rotates in the mean plane of the original nebula. Its plane 
of rotation would be largely influenced by the amount of 
matter that was condensed upon it from any direction and 
the momentum imparted thereby. This will be considered 
further on. 

80. Planets formed at the Perihelion of Cometary Orbits. 
Under the condition that the greater part of the matter that 
was locally condensed, falling sunward from interstellar space 
within the radius of the sun's superior attraction, would fall 
into very elliptical orbits, diffused matter would become much 
more dense near the sun where the perihelia of the projections 
meet. Therefore, if we conceive the sun to be at any period 
in a state represented by any of the spheroidal planetary 
nebulae, so that the nebular density decreased constantly 
from the centre of the sun, such a system might continually 
contract in volume by radiation and yet maintain a similar 
outward state from exterior pneuma projections thereto. 

If the perihelion distance of any such condensation as 
described above fell within the attenuated nebular system 
about the sun, its after projection therefrom would be of less 
velocity, so that it would follow an elliptic orbit of less 
eccentricity. A second perihelion contact with attenuated 
matter about the sun would again act in like manner, so that 
by perihelion resistance matter projected thereto might finally 
fall into a nearly circular orbit. If the resistance were of a 
certain value the above conditions would occur at a single 
perihelion contact. If the resistance at perihelion exceeded 
the momentum required to maintain a nearly circular orbit, 
of about perihelion distance, the projected body would, as 


before stated, drift in spiral lines towards the centre ; or this 
resistance might be partial, so that it might rest in orbital 
equilibrium in a certain internal position and become a part 
of the permanent nebula or planet moving henceforth in a 
nearly circular orbit. If its perihelion of projection glanced 
upon the outer surface of the solar nebula, entering it only 
sufficiently to be deflected from the resistance, the projected 
body might remain nearly in contact. This must, however, 
be a special case. The near comets may have been the result 
of such glancing conditions. The planet Pallas may have 
been formed under similar conditions from a comet that 
glanced upon the limit of the solar nebula sufficiently deep to 
be resisted at its head, until its tail coming forward in the 
same direction condensed about the head and formed a 
nebulous planet, to be henceforth projected in an elliptic 
orbit of small eccentricity. The nebulous planet would 
afterwards condense to its present state by radiation and 
initial gravity. 

81. In the formation of our solar-planetary system, the 
fact should never be lost sight of that the entire planetary 
system is of relatively small mass in comparison with the 
central solar system, being probably not over 1/700 part, 
including all planetary, cometary, and meteoric matter. 
Therefore the parts of the solar-planetary system which 
attained orbital velocity during the nebular condensation at a 
distance from central solar nebula, as a resultant of original 
rotation, acceleration through centralization by gravity, and 
by the perihelion retention of exterior matter here proposed, 
must altogether be taken to form but a small part of the mass 
of the entire system. In the further discussion of the forma- 
tion of the planets this conception of the subject will always 
be inferred. 




82. Energy of the Solar System. It may be held that 
Lord Kelvin has shown demonstratively that the energy of 
the solar system could not, even if it were produced by a 
discrete condensation of cosmic matter, have been maintained 
by this form of condensation throughout the narrowest 
possible limit of past geological time ( 9) *. Therefore we 
have no theory heretofore offered of a condensation system 
by gravity to represent the formation of our solar-planetary 
system with any reasonable probability other than that of 
Laplace and Helmholtz. Nevertheless it is not probable that 
our system was formed by any simple single mechanical 
effect of the action of forces upon surrounding universal 
matter, as generally assumed in special theories, but rather 
that all possible conditions were active that may have con- 
spired to produce the final results. Some of these conditions 
will be now suggested and discussed. 

83. Asymmetry of our Solar-Planetary System. If we sup- 
pose our original nebula throughout its entire volume to have 
been in a uniformly purely gaseous state and of symmetrical 

* Trans. R. S. Edinburgh, vol. xxi. p. 66. 



form, as, for instance, that of a spheroid in revolution we 
should then, no doubt, if the entire system remained gaseous 
until all the planets were consecutively condensed at the 
exterior limit of its nebular equator, according to the theory 
of Laplace, expect to find the planets in symmetrical order of 
distance and of mass. In this case, with proportional time- 
condensation, under the increasing amount of tangential 
impulse due to centralized condensation into gravitation, 
which produces the law of orbit, the distances of the planets 
from the sun and their separate masses would be symme- 
trically proportional, in accordance with the pull of gravitation 
and the tangential momentum of the amount of the con- 
densed matter. 

84. That our solar system does not possess the above 
described symmetry is evident from its formation. We have 
between Jupiter and the earth, particularly, planets in mass 
and density in no way proportional or symmetrical with 
others. The planets exterior and interior to the asteroids, 
taken by themselves alone, have some points of resemblance 
in density and in magnitude. The exterior planets fairly 
resemble one another in number of satellites, assuming that 
Neptune may have several more than the one visible through 
our telescopes ; so that so far as these conditions go we might 
roughly divide the system of planets into two classes. We 
might also possibly make a much more important division, in 
a formative sense by separating them according to the 
direction of rotation, by which the outer planets Uranus and 
Neptune would form a class by themselves. Notwithstanding 
all these minor classified resemblances, there evidently 
remains beyond this a general want of symmetry in the entire 
solar-planetary system. Therefore, if we conceive an original 
uniform gaseous or a generally symmetrical system for our 
primitive nebula placed under the condition of uniform 
condensation, according to the theory of Laplace, we must 
conclude that this system has experienced material modi- 


fications during the period of condensation to form our 
planetary system. This will be now considered by taking at 
first as groundwork the effects of the condensation of a purely 
symmetrical nebula, and afterwards suggesting what modifi- 
cations there may have been in some parts of the system. 

85. A Symmetrical Gaseous Solar- Planetary System. We 
may take our solar system in its nebular state at a certain 
period to be of the simplest construction shown diagrammati- 
cally by fig. 9, which is intended to represent a very oblate 
spheroid in revolution upon its symmetrical axis, a a . We 

Fig. 9. 

may assume that the whole spheroidal surface of the nebula 
would be radiating heat equally per unit of area into space. 
The volume of nebula in a gaseous state would therefore be 
limited at any time by the quantity of matter that could be 
maintained in this state, by the initial heat of the primitive 
gaseous system, together with the amount of heat given out 
from the centre, where there would be a sun-forming con- 

86. In the above-stated case the important heat-maintaining 
centre or incipient sun would react through its condensation 
as a heat-radiator to the surrounding nebula, and disperse its 
heat due to condensation in proportion inversely to the square 
of the distance from the incipient sun-surface. The heat- 
radiants being therefore equal at equal distances from this 
sun, would tend to maintain a circumscribed globe of nebula 


in a gaseous state, if its exterior temperature was falling. 
We may assume that this inscribed imaginary hotter globe, 
fulfilling the condition of equal radiants, would be so large 
that its surface could just be inscribed in the oblate nebulous 
spheroid which we now take for our complete nebula, as 
shown by the inner circle of the figure. 

It is readily seen that matter placed outside the theoretical 
globe suggested above would be more rapidly lowering its 
temperature by radiation, from its greater extent of surface 
in proportion to its depth of volume, than the matter of the 
inscribed globe. At the same time this external matter would 
be receiving less heat from the sun-forming centre. The 
peripheral matter in the general revolution of the system 
would also possess its highest tangential velocity in the 
equatorial plane. 

The whole of these conditions, upon loss of heat of the 
system by contraction through radiation, would cause a stress 
at a certain nearly cylindrical plane parallel to the axis of 
rotation, wherein internal or sun-forming matter would, by 
continuous cohesion or gravitation, in the gaseous state, be 
drawing away from the peripheral zone, which was moving 
at higher tangential velocity. This zone, by continuity of 
radiation of its own heat after its detachment, would be 
induced to contract by condensation upon itself. 

87. In this manner, if we assume peripheral matter to be 
proportionally distributed about the axis of revolution of a 
spheroidal nebula, the stress-plane would be that of the 
separation of the peripheral .matter, which might then form a 
detached zone or ring upon the theoretical conditions given 
by Laplace. The extent or distance of the ring from the sun 
would depend upon the extent of the globe of centre-tending 
condensation maintained by the central heat at the time. 
That is, really upon the contraction at the surface of such a 
globe as herein imagined, by loss of its heat through 


In the separated ring or planet-zone, as here suggested, 
we presume a perfectly equal distribution of matter about the 
equatorial regions of the oblate spheroidal nebula ; but as we 
take the conditions of radiants from the sun- centre, the same 
principles would approximately hold if the ring were more or 
less incomplete or denser in any part. The accident of a per- 
fect ring or system of rings which occurs around Saturn may 
be conceived possibly to be a unique phenomenon of the nearly 
equal distribution of cosmic matter about a gravitation-centre. 

88. The principles here discussed are a possible explanation 
of those given by Laplace, under which the nebula forming 
the exterior planets would be consecutively abandoned. 
Whether there may have been some modifications of this 
during the condensation of our own planetary system will be 
discussed hereafter, but to follow our theme, we will take a 
uniformly distributed nebular system in revolution as being 
of spheroidal form, in which case the nebula might con- 
tinuously go through the same set of changes as consecutive 
zones or planetary systems were detached from the central 
sun-system. Thus, assume (fig. 9) aba' b f to represent 
diagrammatically the nebulous spheroid in section with axis 
a a! and equatorial plane bb f . The sun-forming central 
matter under condensation, at a certain period maintaining 
equal heat-radiants, would circumscribe S. The zone of 
planet-forming matter supposed to be moving at orbital 
velocity, upon condensation of the sun to a certain extent 
would have its annular centre of gravitation circumscribed 
about p p r . The lateral exterior matter X, #, #, #, upon 
condensation would fall upon the sun or the planetary zone 
in gravitation-equation. 

89. It will be seen that the zone-ring pp r , although de- 
tached, would still for a long time maintain its condensing 
condition, so that the central axis of the section of the ring 
might by condensation of surrounding matter become in time 
incandescent, or hotter than the nebulous sun. Therefore, 


the sun in the interior of the equatorial orbit-plane of this 
zone b b' would, through heat-exchanges, not suffer much 
loss of heat, whereas at the same time it would be radiating 
heat freely about its polar surface a a' constantly into space to 
cause the sun's contraction in this polar direction. Further, 
after separation of a ring, during general contraction of the 
now separated central nebula its equatorial parts would keep 
up its extension in this direction by tangential momentum, 
and form a closed system by the attraction of matter towards 
the axis of the ring, which, as before stated, would become 
heated in proportion to the activity of its condensation. 

90. Under the conditions proposed above, heat could be 
freely radiated from the polar regions into space to cause the 
contraction of the nebular sun in this direction ; but this 
could not occur about the equatorial regions, where heat- 
exchanges with the condensing zone would prevent free 
radiation into space. The difference of local contraction- 
areas upon the surface of the solar nebula, in conjunction 
with the tangential momentum of its peripheral parts, would 
cause it to return to its oblate spheroidal form but of smaller 
volume. The spheroidal form would also influence gravi- 
tating matter to fall towards . the plane of the solar equator, 
as suggested originally by Kant and shown more definitely 
by Newcomb for attraction toward an oblate spheroid *. 
All these conditions show that after separation of a zone-ring 
the system would again become oblate and in a condition 
to separate another planet-zone therefrom about its equator 
upon the same principles. 

91. It will be seen that the surfaces of the sun and the 
planet-zone that would conserve their heat, therefore their 
nebular condition, would be those parts directly opposite 
to each other. Whereas the outer periphery of the detached 
zone-ring would be freely radiating its heat into open space, 

* l Popular Astronomy/ p. 513. 


so that the contraction of the zone-ring upon itself by con- 
densation would be upon its outward and lateral parts only, 
the inner parts retaining constant nebulosity by exchanges of 
heat with the nebulous sun's surface, as before stated. The 
contraction of the condensed outer matter would also be 
directed sunward by gravity, assuming the outer parts of 
the ring were moving at less than orbital velocity, which is 
necessary for the concentration of the system. 

Accepting the theoretical matter for the case proposed 
above, if we take the centralizing globular nebular system of 
the sun at the formation of the Neptune-ring or planet- 
system, the diameter of the imaginary central globe at its 
full limits would then exceed the diameter of the orbit of 
Uranus ; at the formation of Uranus it would exceed that of 
Saturn ; at the formation of Saturn it would exceed that of 
Jupiter ; and so on. 

92. Modes of Condensation of Interior Planets in a Sphe- 
roidal Nebular System. If we take the earliest conditions, 
the outer planets might very well go through the same 
transitions of the nebula as suggested above. We may 
observe, however, under the conditions proposed, that heat 
being always best maintained in the equatorial plane by 
exchanges with the newly formed planet-rings, where also 
the tangential impulse would be greatest, and the greater 
condensations going on constantly at right angles to this 
plane, where it was open to free radiation of heat, our 
assumed spheroid must constantly and rapidly flatten out its 
form of section in the direction of the orbit-plane of the zone- 
ring. Therefore the early planets, say Jupiter and those 
exterior to it, might be condensed from a spheroidal nebula 
under purely nebular conditions ; whereas the inner planets 
assumed to be formed afterwards when the spheroidal solar 
nebula had become much flattened, and thereby presented 
much greater extent of radiation surface to depth of volume 
about the equatorial zone, may have sunk in temperature, so 


as to produce exterior local condensation within the boun- 
daries of the nebula. Under these conditions the continuity 
of the former nebular state would be wholly or partially 
changed. So that the planetary zone-system of Jupiter and 
its superior planets might somewhat resemble the condensation 
of our sun, but the inner planetary system might be wholly 
or partially formed from a condensation which took the 
primitive form of discrete or meteoric matter. This would 
partially account for differences of density, of rotation, and 
some other conditions which will be more fully considered 
hereafter upon discussion of certain exterior conditions. 

93. Mode of Condensation of the Extreme Outer Solar 
Nebula. In this we may possibly again find a modification 
or want of continuity of nebular conditions in relation to the 
planets Uranus and Neptune, which may be inferred from 
the direction of revolution of their satellites, under conditions 
already stated. In the case of these planets, from the want of 
concentrative force in the original nebula through weakness 
of effective gravitation due to distance, there would be a 
weaker centering-tendency of the peripheral matter. There 
would also be less heat radiated inversely proportional to the 
distance of the central heat-focus of the condensing sun. The 
tangential impulse being assumed sufficient to maintain the 
orbit of a zone-ring, local condensations might in this case 
occur at first to discrete matter, and the nebular system 
would thereby largely disappear. We may assume a gaseous 
system so extensive that the radiation capable of producing 
condensation would act superficially upon matter within a 
limited depth only of the nebula ; so that gravitation would 
possess insufficient energy to draw this matter, if possessed 
of less than orbital tangential momentum in a gaseous state 
than that which would maintain it at its radial orbit, through 
the resistance of the interior nebula from the boundaries of a 
system so vast as that of Uranus or Neptune, under certain 
conditions suggested ( 13) and others to be discussed. This 


would produce a reverse direction of rotation, as will be shown 

94. The Breaking-up of Gaseous Zone-Systems. The 
perfect state of equilibrium of nebular matter necessary for 
the complete formation of a planet-ring or zone (fig. $,pp'), 
if such ever existed, except among satellites, could scarcely 
remain for a long time, as a slight disturbing cause at any 
position throughout the extensive orbit would destroy this 
equilibrium in such a manner as to permit the gravitation of 
its own mass to draw its parts together into the only form of 
static equilibrium that could be established, that is, a globe. 
Further, as before stated, in early planetary stages the intru- 
sion into the solar nebula of exterior matter possessed of 
sufficient eccentricity to bring it at perihelion within the 
planet-ring orbit, would cause its inclusion within this ring 
owing to the resistance of the nebular matter of the ring 
itself to the continuity of its projection. In this manner all 
comets exterior to the system, or meteorites of great eccen- 
tricity, would be absorbed if brought in orbit- contact with 
the planet-ring. And although we may assume that the 
mass of a comet or of a shower of meteorites projected in a 
cometary orbit might not materially disturb the mean orbit 
of the ring-system, it might upon its intrusion possess quite 
sufficient momentum to destroy the equilibrium of a perfect 
ring, if such existed. This would not only be caused by 
its mass, but also by the local heat engendered, and the 
elastic expansion it would cause near the place of intru- 
sion, together with the local drifting force due to differ- 
ence of velocity and inclination of orbit between the orbits 
of the planet-ring and the intruded cometic or meteoric 

There is no doubt that if a planet-ring were perfectly 
symmetrical the inner attraction of its parts in a gaseous 
state might, under contraction through radiation from its 
own condensing matter, reduce its section until it might 


form even a liquid ring or rings ; and this or these might 
again be detached into beaded strings of satellites, a condition 
which possibly holds in the case of Saturn's rings. A ring 
of perfect symmetrical condensation might finally part in one 
place only, and form a single satellite by its matter being 
drawn together by gravitation. This was possibly the case 
with our own moon, as will be discussed later on ; but such 
perfect equilibrium of distribution of matter surrounding a 
gravitation centre could scarcely hold to the extent of a 
planetary orbit about the sup, and the zonal abandonment 
principle of Laplace is maintained if the zone is even imper- 
fect in its circumference. 

95. Influencing Condition in Periods of Planet-formation. 
Critical Temperatures. Noting the irregularity of the masses 
of the planets, which cannot be accounted for by proportional 
condensations in time, there were no doubt present special 
inducing causes active for the time only, by which we may 
assume a greater or smaller nebulous zone-system was 
detached at any particular period from the central solar 
system. One of these causes was most probably the effect 
produced at certain times by the rapid condensation of 
nebulous matter to a liquid state at its critical temperature, 
within parts of the solar nebula, the mass of which we assume 
to be moving at less than orbital velocity at its periphery. 
Such condensed matter would, in the outer parts of the 
system, be immediately precipitated nearer towards the sun. 
This might not, as an early condition, be wholly possible 
with dissociated matter, that would only condense to a gas, 
which might suffer resistance from the elasticity of the 
highly heated interior. It might have occurred after the 
formation of the larger planets upon the uniform cooling of 
the entire system. If the temperature of the nebula was 
partially reduced at any period so as to cause it to pass from 
a gaseous to a vapourous state at any radius within the solar 
nebula, it is certain that the denser or metallic matter so 


reduced to vapour would condense suddenly at its critical 
point, by a very slight further depression of temperature, to 
the liquid form. The general equable state of temperature of 
the nebula might permit, for instance, certain metals in the 
vapourous state to occupy large volumes, and to condense 
afterwards with a very slight depression of temperature at 
the critical point, causing a sudden interior collapse, and 
thereby the separation of a volume of peripheral matter 
which would afterwards maintain an orbit position consistent 
with its initial tangential impulse. These condensations 
might be at any distance from the sun within the nearly 
transparent nebula, according to the density and vapour 
temperature of the special element that was condensed. 
The amount of nebular matter, whether very voluminous or 
not, separated by the tension of an internal condensation and 
moving at about orbital rate at any time would after detach- 
ment maintain its free orbital position. 

96. The interior of the zone-ring of detached matter would 
be the stress-plane of the exterior part of the critical conden- 
sation. Whether the matter condensed at its critical point 
remained as vapourous cloud in its precipitation towards or 
about the sun afterwards would depend upon the reaction of 
the heat of its condensation, radiated from the central system 
or sun at the time. 

97. Upon the above-stated conditions, as far as they go, it 
is seen that the condensation of matter at the critical point 
would produce a permanent strain within the nebula, so that, 
seeing the nature of chemical elements and their very varied 
critical temperatures, the separation of planet-rings from the 
central solar system is not necessarily a uniform process 
depending upon a continuous condensation as proposed by 
Laplace. By condensation of interior matter at its critical 
point of temperature, a planet-forming zone or system may, 
upon this hypothesis, be separated from the sun at a certain 
time ; and for a long period after this separation the sun may 



slowly condense nebulous matter upon itself only ; until 
again, under certain conditions of critical temperature of the 
materials of the nebula, another planet-zone or system may 
be detached. Therefore, owing to the great differences in the 
critical temperatures of known matter, the zone or volume of 
detached matter may be of relatively large or small mass, and 
the planet formed therefrom will be consistent with this so 
far as the principle in question is active. 

98. It will be observed that the condensation of any 
refractory metal from its vapourous state within the nebula 
would affect this particular metal only, and the vapours of 
other less refractory matter would remain in a nebulous state. 
Therefore, in any rotary nebular condensation, the denser, 
more refractory matter, moving at equal angular velocity, 
with the peripheral matter moving at nearly orbital velocity, 
must always drift to an inner position in spiral lines, being 
accelerated by gravitation also thereto. If the condensation 
remained, it would come to rest only as regards centralization 
when it attained an orbital position. Under these conditions 
internal planets must be formed of denser, more refractory 
matter than external ones. 

A particular case of critical condensation would be one in 
which oxygen and hydrogen in a mixed state, below the 
temperature at which they must remain in contact permanent 
gases, were united into vapour. This condensation might be 
caused by a discharge of electricity, from an incidental 
chemical combination of prevalent elements within the 
nebular system. 

99. The condensation to cloud, metallic or other, at the 
critical temperature of any annulus of nebula at a distance 
from the sun less than the radius of the condensing peripheral 
zone would obscure the exterior matter from radiation of the 
more highly heated central system, as the central heat and 
light would be reflected back from the condensed particles. 
This would cause the more rapid condensation of the detached 


zone, whose tangential momentum would prevent its conden- 
sation upon the sun. This clouding effect would be repro- 
duced at any following critical condensation of the matter of 
the sun system, and would again tend to condense a detached 
planet-zone or system ; but it is not proposed that such a 
form of condensation is alone active in our planetary system. 
Other conditions have been already suggested, and will be 
further considered. 

100. It may be noted under the conditions given above that 
a period when an element was under condensation at its 
critical point about the sun would be a period when his 
radiation would be materially obstructed. Therefore when 
an outer planet would receive much less of his heat. Such a 
period, which may in some cases have lasted many thousands 
of years for the complete condensation of a single element, 
may in recent geological time have produced a glacial period 
upon a wide extent of the earth, under certain conditions of 
distribution of land areas and direction of oceanic currents, 
which I have previously considered for geologically recent 
glacial epochs *. 

101. Modifying Conditions. In the construction given 
( 76) we assumed that an entire nebular ring, extending 
possibly within 10 from the sun on each side of the plane of 
the earth's orbit, was condensed to form the planet. This, 
however, we may presume was not the case. Very probably, 
as before suggested, the planet-forming ring was never 
perfect, or if perfect it is improbable that it should have 
condensed entirely at once into a single planet. Possibly 
condensations to meteorites form a common factor when the 
nebular system sinks below a certain critical temperature. 
Therefore, the earth's nebular ring might split up into a 
single planet and satellite upon one side of its orbit and be 
distributed in meteorites upon the other side. If these 

* British Association "Reports, 1885, p. 1020. 


meteorites were of slightly different orbit-period from the 
earth they would finally unite with it at conjunction^ but if 
of the same period they would maintain their vis viva and 
not be detected by any calculation in the variation of the 
earth's mean course or by telescopic observation. That is, 
assuming such meteorites to resemble those that fall upon 
the earth, which may have fallen from outer planet-rings, 
and which are generally of masses not exceeding a few 
hundred pounds. 

102. The greatest condensation to form a planet would not 
in a uniform density-ring be at an intermediate position 
between an inner and outer planet, as may be inferred from 
the diagram fig. 9. The gaseous state would be best main- 
tained towards the interior of the ring by heat exchanged 
with the nebulous sun. The condensation would, therefore, 
be on the outer surface of the ring at the greatest distance 
from the sun, where heat could be freely radiated into space, 
as before stated ( 89). This exterior condensing matter, if 
it fell towards the sun, would cause the orbit-position of the 
new-forming planet to be towards the interior of the ring. 
In this position the precipitation of exterior matter falling in 
spiral path would give excess of velocity from gravitation to 
the interior matter beyond its original angular velocity, and 
might set the planet in gravitation equilibrium for a nearly 
circular orbit and in rotation at this inner position, although 
the original nebula had less angular velocity, as will be 
shown that it may have had further on. It is not necessary, 
therefore, upon the principle of exterior radiation to assume 
that a planet was formed entirely of matter of a nebulous 
planet-zone ; it is much more probable that the inner con- 
densations were at first upon the nebulous sun's surface, 
and did not separate therefrom until a dense motive system 
had been already formed in one position. Neither is it 
necessary to assume in all cases a single ring or a perfect 
ring-system; there may have been many imperfect or partial 


rings detached before these formed a single planet, these 
being united afterwards by variation of time-orbits, and 
crossings of perihelia through eccentricity, or be drifting in 
spiral lines inwards. 

The voluminous nebula of Jupiter would affect the nebular 
conditions in the formation of an inner planet. This will 
be best considered in relation to the formation of inferior 
planets, the Asteroids, and Mars, and the possible effects of 
this nebula upon the formation of the Earth. The subject 
will therefore be deferred. 

[ 80 ] 



103. The Distances of the Planets from the Sun appear to 
be in somewhat symmetrical order in individual distribution 
of position, although their masses do not indicate any law 
for their formation consistent with the condensation of a 
gaseous system or of a uniformly distributed discrete system 
by a decrease of density outward from the gravitation centre 
or sun. The approximately symmetrical order of distances, 
without relation to the amount of distribution of matter, was 
pointed out by Titius in 1772 *, which became known as 
Bode's Law owing to the special attention called to it by that 
astronomer f. It is illustrated in the following table, the 
scale of measurement being the sun to earth unit. 

Table of Bode's Law. 






J up. Sat. 













B. eq. . 




















Obs. ... 


72 1 







* See Miss Clerke's ; History of Astronomy in the 19th Century/ p. 87. 
t W. T. Lynn, ' Observatory,' vol. xvi. (April, 1893) p. 178. 


In the first line 4- '4 is given as an arbitrary plus constant. 
It may be noticed that it is correct to observation, according 
to the law, for the places of the Earth and Jupiter, irregular 
with the inferior planets and Mars, and should be omitted 
altogether as a plus constant for the outer planets, failing 
entirely for Neptune. The second line in the table is in 
geometrical series from Venus, which is quite arbitrarily 
taken as 3. The third line gives the theoretical deduction 
of Bode's Law. The fourth line gives approximately the true 
distance as found by observation. 

With Uranus and Neptune, there appears to be a certain 
element of orbital time relations ; the year of Neptune being 
about double that of Uranus. 

104. The Masses of the Planets. No law has been dis- 
covered for comparison of the masses of the planets, except 
that the four inner planets are smaller and have a mean 
density more than five times greater than that of the four 
outer ones, which may indicate that their formation has been 
upon a different plan. In the outer planets there is a kind 
of proportion of masses to spaces, which agrees approximately 
with an assumed decrement of density of nebulous matter 
employed in their formation not inconsistent with the manner 
in which mixed gaseous matter would probably condense 
when placed around a gravitation centre. 

105. Following the demonstrations of the theory of Laplace, 
the density of matter to form the planet may be estimated by 
reversing the process of its condensation ; that is, by the 
dissipation of the mass of the planet into the assumed original 
nebular zone-ring volume it formerly occupied. Assuming 
that the planet will be formed upon the inner surface of the 
ring, as stated above, since radiation, and therefore contraction, 
must be exterior to this, we may for calculation take the mean 
distances from the sun of any pair of planets and make half 
their difference the radius of the ring assumed for argument 
of circular section. Calling this r, the section of the ring 



will be Trr 2 ; making TI the mean radius of the orbit of the ring 
and therefore its circumference 2 < 7rr 1 , we have for the volume 
of the ring 27r 2 r JJ r 1 . We may conceive the ring of a certain 
oblateness ; say this diminishes the area of the section by 
two thirds, our formula then becomes f TrV 2 ^ for the corrected 
ring volume. Now taking the planet as a sphere f7rr 2 3 , 
y 2 being its radius, and dividing this into the volume of the 
ring, we obtain its assumed original density in the planet's 
specific density units. For comparison it is convenient to 
make the unit of density that of air at the earth's surface. 
Then the specific density of the earth is found by multiplying 
its volume by 5' 6 its specific density compared with water 
and 800 the ratio of air to water. Calling this m, the com- 
plete formula becomes 

Other planets may be taken in a similar manner, changing 
m to m according to the data for the density of the planet. 
The following table is taken from the above formula, adopting 
Bode's law for the mean place of the Asteroids : 

Table of Proportions of Densities of tlie Nebular Planet 
Ring-spaces to the Density of Air. 

Mercury .... 1/1,060,000,000 

Venus 1/81,230,000 

Earth 1/372,400,400 

Mars 1/418,800,000,000 

Jupiter .... 1/505,900,000 

Saturn 1/15,960,000,000 

Uranus 1/248,400,000,000 

Neptune . . (?) 1/5,000,000,000,000 

106. The irregularity of these figures appears to indicate 
the improbability of the part of our nebula which formed the 
planets having been condensed from matter symmetrically 
distributed in an oblate spheroidal form, although perhaps 
this may not be altogether inconsistent as a form of condensa- 
tion of the four outer planets and their satellites, as the table 
shows great increase of tenuity outwardly. 


107. The decrease of the former nebular density-space 
between Jupiter and Saturn and Saturn and Uranus varies 
about as the powers of 1*17 to 1'12, and possibly if we knew 
its extent of original nebula, Neptune might follow a similar 
ratio. The figures appear to show upon the ring-nebula 
hypothesis that Uranus was condensed from matter weighing 
only about 4*6 grains to the square mile, a degree of tenuity 
difficult to imagine in a concrete system. This extreme 
tenuity might, however, be greatly reduced, if, instead of 
taking the oblate spheroid form before proposed we were to 
assume a more lenticular shape for our planet-forming nebula, 
thinning-out to a nearly flat plane, which would be quite 
consistent with observed forms of nebulae seen in the heavens, 
of which we are supposed to observe the thinnest section *. 
Under this condition, the nebula might not vary much in 
density in the outer series of planets from Jupiter to Neptune 

108. Taking the oblate spheroidal form of nebula assumed 
for our solar nebula immediately before its condensation to 
planets, we may also suppose that effects of condensation 
through surface radiation produced some differences. The 
extensive and attenuated plane of Neptune and Uranus being 
open to this surface radiation would cause the nebula about 
the positions of these outer planets partially to condense into 
solid matter before the large masses of Saturn and Jupiter 
had considerably changed. Matter so formed and so widely 
distributed as it probably was originally before the formation 
of these outer planets, would scarcely condense entirely into a 
cohesive gaseous system, but we might more probably have 
at an early stage the formation of a motive discrete system 
composed of minute minor local condensation or of dust in 
the orbit-plane, according to the system of Kant and Faye. 
Such a system could only unite to form a planet if the original 

* IjF I. 200 Leonis Minoris, Plate II. c ; $ I. 53 Pegasi, &c. 



angular velocity of the particles were less than the tangential 
velocity of a particle in gravitation equilibrium according to 
the law of orbit, so that the particles separately condensed 
would fall in the direction of the nebulous sun of the period 
into elliptical 'orbits. Such particles would be of different 
orbit periods from differences of distance at the points of con- 
densation, so that they might afterwards come into collision 
at the crossing of orbits about the coincident orbit position to 
cohere and form a planet ; or they might unite in the exterior 
of the central nebulous system at the perihelia of their pro- 
jection, which would henceforth become their orbit. This 
will be further discussed. 

109. Probable Form of the Original Planetary Nebula. 
If we take the original nebula to have been of about the 
same density at periods when the separate planets were 
abandoned, we may then plot a section that would represent 
the form which the planet-forming nebula would assume after 
a certain amount of condensation. This would show a con- 
siderable rounded projection over the positions of the planets 
Jupiter and Saturn, and as the Sun would also be contracting, 
this part of the nebula of the solar system would for a period 
assume a convoluted discoid form, possibly as represented in 
fig. 10 ; the positions of Saturn and Jupiter being shown at 
S, J. In the present construction we are considering the 
planetary forming nebula only which will be about the orbit- 
plane, and omitting all consideration of the circumscribing 
sun-forming pneuma system, which would maintain the 
spheroidal form. 

110. In accepting this form, it is extremely probable that 
there was some external cause for the exceptionally large 
masses of matter in Jupiter and Saturn. Possibly these 
planets represent factors of an early intrusion of a local con- 
densation of the nebula that formed at a distance from the 
solar plane, which was afterwards projected into it by attrac- 
tion to the sun. Such projections might be constrained to 


follow a circular orbit by the resistance they would encounter 
if they entered the sun's nebula at about the perihelia of their 
projections thereto. As before suggested, representation of a 
part of such a form of nebula seen in plan upon a large scale 
may possibly be found in the great nebula of Andromeda, 
Plate II., #, or in a more pronounced form in the ring-nebula 
of Lyra, Plate II., g. In fig. 10 matter is shown distributed 

Fig. 10. 

symmetrically about the orbit-plane ; but it is more probable, 
as will be seen later on, that the planet-forming disc was not 
of this symmetrical form, but irregularly convoluted while 
still in a nebulous state. 

111. Effects of the voluminous Ring of Jupiter here proposed 
upon the intra-Jupiter Solar System. In assuming an ex- 
tensive zone-ring of nebulous matter for the formation of 
Jupiter, the conditions that will be presented for interior 
planetary formation become materially modified from the 
uniformity of direction-gravitation due to the sun only, as 
before proposed. In the case before us, the intervening- 
interior nebulous matter towards the Jupiter-ring would be 
solicited by two unequal gravitation systems, that of the sun 
and that of Jupiter ; the Jupiter ring being practically active 
near its surface only. Assuming the intervening matter 
between this ring and the sun at the time to be wholly 
nebulous, this nebular condition being assumed to be largely 
maintained by the heat of the interior of the ring, there 
would then be a strong tendency through cohesion for the 
nebulous matter near this ring to be drawn either towards 
the sun or towards Jupiter in condensation. This would 


occur particularly from the condensation due to radiation 
being greatest in relation to the depth of volume in the most 
attenuated part of the nebula, fig. 10, A. Therefore in this 
case the condensation would be from the surface in the orbit- 
plane of the nebula and would tend to break it up into separate 
superficial small local condensations. Further, from the great 
extension assumed for the nebular system of Jupiter exterior 
to the plane of orbit and the large volume of the nebulous 
sun at the time, there would be little excess of attraction 
towards the plane of orbit for nebulous matter within the 
orbit of Jupiter. Therefore, upon condensation of such a 
nebula, owing to the equilibrium of its position between the 
attractions of the Sun and of Jupiter, there would be a local 
tendency to form very small condensations or planets, par- 
ticularly near the interior of the extensive nebular ring of 
Jupiter. These are probably invisible from our distance. 
Such small planets, at least the earlier ones formed upon the 
exterior radiating surface, would possess great inclination to 
the plane of the orbit of Jupiter in moving as free bodies 
under the stronger influence of the sun's attraction. In the 
equilibrium of position of condensation we have possibly -the 
principal reason that the Asteroids are of small mass, and 
particularly that these small bodies should often be found 
moving in eccentric and inclined orbits, omitting extreme 
cases of inclinations probably due in part to other causes 
previously considered ( 80). 

112. Relative Rate of Cooling of the Intra-Jupiter Sun- 
system. After the sun's spheroidal volume had retired from 
the inner surface of the ring of Jupiter's nebula, and this 
nebula was becoming of insufficient temperature to maintain 
an inner nebulous system, and many Asteroids of inclined 
plane of orbit had been condensed, the remaining nebula may 
have been disposed approximately as represented in fig. 10, 
where the letters S, J, A, M, E represent diagrammatically 
the positions of Saturn, Jupiter, the Asteroids, Mars, and the 


Earth. Taking this outline of the section of the nebula and 
assuming surface radiation to be equal from all superficies, 
it will be seen that the narrow neck M A of the nebula of 
Mars and the Asteroids would by its shallow depth be the 
earlier part to cool down to condensation point, leaving Jupiter 
and Saturn still in a heated nebular condition. This condition 
of radiation remaining constant, the Asteroids, Mars, and our 
Earth would certainly be condensed to a liquid state long 
before the large mass of Jupiter could have lost its nebular 
condition. It therefore becomes probable that the condensation 
of Mars and of our Earth may have taken place much earlier 
or certainly not later than that of Saturn, so that our planet 
in a liquid or solid state would be much older than Jupiter. 
It is even probable that Jupiter may be considered only as an 
advanced minor solar system, and that at the present time it 
may not have its surface about its equator condensed to a 
solid coating : this will be discussed further on. 

113. After the cooling of Mars and the Earth to a liquid 
or solid mass, the consecutive condensations of the inferior 
system of Venus and afterwards Mercury would again possibly 
more nearly fall into a system of periodic condensations similar 
to the earlier condensations of the outer planets, following 
the plan of consecutive exterior condensations as proposed by 
Laplace, except that with the inner planets we may have had 
a denser medium, or more probably a medium pervaded by 
discrete matter from which these planets were condensed, 
which would account for their superior densities and rotation 
periods subjects that will be fully considered further on. 



114. The Direction of the Solar Axis of Rotation. Let 
s f s ff (fig. 11) represent the nearest stars to our sun c, and the 
line of direction s r c s ff therefore the mean gravitation plane 
of the planets' orbits as before defined ( 79). Let s, s, s, s 
be other stars more distant from the sun than s' s /f . Bisect 

Fig. 11. 



the distances between the sun and each of the stars, supposed 
of equal mass with the sun, by the arcs shown. The dotted 
lines embracing these bisections would represent the extent 
of the sun's original pneuma system, or that of the mean 
gravitation influences between the sun and these stars. 


Assume the solar pneuma in revolution, when free from 
surrounding matter through condensation, with its axis at 
right angles to the plane s' s". Then it is clear that a larger 
amount of matter in further condensation would fall to 
the sun in the directions a a! than in the directions b b f ; 
and as this matter would in condensation carry with it the 
original tangential momentum of the uniform angular velocity 
of the pneuma system, the equatorial plane would thereby be 
elevated from the original solar plane s' s" towards a a', 
the direction from which the greatest amount of gravitating 
matter would fall ; so that the orbit-plane of planet- forming 
matter would be s' s", but the momentum of the sun's con- 
densation greatest in the plane a a' and its equator be subject 
to the combined directive influences of s s" and a a 1 . The 
plane of the stars taken above for illustration must be under- 
stood to be purely diagrammatical ; no such plane exists, 
neither could the axis of original pneuma rotation be defined, 
the plane produced would be undulatory, but the principle 
will hold for stars in any direction from the sun and any 
plane of rotation. Nebulae that approach the spheroidal form 
of which the great nebula in Andromeda may be taken as a 
type (Plate II., a) are generally asymmetrical, being more 
dense in certain directions *. Such systems in condensation 
would not therefore produce axes symmetrical in the plane ot 

115. Direction of the Planets' Axes. If the suggested 
discoid form of the planet-forming nebula (fig. 10) existed 
with matter symmetrically distributed about the equatorial 
plane, the planetary axis would be vertical to the plane, but it 
is not at all necessary to assume the planet-forming matter 
to be placed symmetrically. If it were not so placed, the 
direction of rotation of a planet would be in equation with 
the mean momentum of matter falling to the planetary 

* y I. 200, tf I. 205, tf I. 53, M 81. 


centre from any direction. Further, the pneuma system is 
assumed to extend in all directions and that condensations 
occur at its outer surface ; therefore condensations would fall 
into the planetary plane in all directions, such matter being 
projected in cometary orbits. Matter thus projected towards 
the sun and towards the denser equatorial nebular plane 
would be absorbed in the nebular system and give local 
directive influences by causing intermotion within the nebula, 
without necessarily displacing greatly the positions of the 
planetary condensations then forming. This assumes the 
planet to be of much greater mass than the units of intruded 
matter falling constantly from various directions, and there- 
fore such intrusions to be of insufficient momentum individually 
to materially disturb its orbital position. The projection 
of eccentric cometary matter would be very much more 
frequent in early times than at present, as all such projections 
would become absorbed in the solar-planetary nebular system. 

Fig. 12. 

We must assume, from the extent of a pneuma-solar 
system, that cometary matter from exterior condensation 
would materially affect the intermotion of the parts of any 
system of nebula that was in a state suitable for planetary 
condensation, wherein every planet would form a gravitation 
centre with directive influences. Such intrusions only by 
directive impression according to the momentum of matter 
projected from any direction might disturb the mean plane of 
orbit of the planet, or induce obliquity of axis by diverting 
the mean revolution direction in the condensing nebula of 
which the planet was being formed. Under these conditions 
the theoretical general half section of the planet-forming 


nebula given in fig. 10 might be changed at the point of the 
commencement of planetary condensation to a section more 
nearly as represented by fig. 12 for the portions of the nebula 
about Jupiter and Saturn employed in their formation. 

116. Rotation of the Sun. As long as a spheroidal solar 
system of pneuma could remain in a state of gaseous 
cohesion entirely by the effects of its own elasticity through 
expansion by internal heat, so as to divert the direct 
action of free gravitation of its outer parts into a pressure 
upon the inner parts, such a system revolving upon its 
symmetrical axis in a frictionless medium or a vacuum 
would revolve entire. It would also revolve at equal angular 
velocity in all its parts as the least frictional form of motion 
for a 'continuous pneuma or highly gaseous system. In this 
case the peripheral velocity of the entire pneuma or its more 
condensed nebula must be such as will permit the exterior 
part of the nebulous matter to remain in contact or in 
cohesion upon the extreme outer surface of the entire system. 
This entails that the velocity of rotation of any part of the 
system must not exceed that which produces gravitation 
equilibrium according to the law of orbit. If the peripheral 
matter exceeded this orbital velocity at any time, it would be 
thrown off and depart from the central system. If the 
pneuma moved with a smaller velocity, it would press the 
gaseous matter towards the centre, forming a nebular system, 
and attain thereby a higher velocity by gravitation until 
the peripheral matter, maintaining its original angular velocity 
with the excess due to gravitation in falling towards the 
centre, had attained an orbital velocity, and then its pressure 
would cease. 

117. Assuming such a nebular system as above defined, 
free from surrounding exterior attraction, moving at such a 
velocity that its peripheral parts could not rest in gravitation 
equilibrium as free particles, but must exert a pressure upon 
the system, then from any loss of heat or of internal elastic 


force the tendency of all exterior matter would be active to 
press forward towards the centre, as the virtual velocity 
would be less in any interior part than that which would 
separately maintain its matter at the original distance of 
radius in a free state of orbital motion. Upon this con- 
dition the surrounding pressure within a gravitation system, 
if it exceeded the elastic force of the internal heat of 
the system for materials in a highly gaseous state, would 
form a dense mass or sun in the central part. In forming 
this central dense condensation, omitting the friction of the 
system which would produce heat, or assuming this equal 
to the acceleration of its gravity in falling sunward, we may 
take it that all matter condensed upon the sun would carry 
with it the linear velocity of its former position, which would 
be greater in the proportion of its original linear circum- 
ference to the circumference of the sun upon which it was 
afterwards condensed. This condensed matter would there- 
fore rotate the sun in proportion to the mean excess of linear 
velocity or momentum of the matter condensed upon it from 
which it was formed. 

118. To demonstrate the above proposition, we may assume 
that all nebulous matter formerly condensed upon the sun in 
consecutive shell-layers over its surface, at first from directive 
pressure of the interior parts of the nebula surrounding it, 
which we assume was moving in mass at equal angular 
velocity with the centre or sun. Then the condensations 
from the interior parts would impress small excess of linear 
velocity upon the sun in the early or central shells of con- 
densation, and the more exterior parts consecutively higher 
velocity from the condensation of these more distant parts 
of the nebula in the outer shells, which theoretically would 
be consecutively brought down to the surface of the sun. 
So that with increasing volume the condensed central sun 
would attain constantly increasing velocity of rotation. 

By condensation in the above paragraph, is intended purely 


gaseous condensation, the condensation being due to loss of 
heat by radiation from the system by which the whole mass 
maintained the elastic force with which it formerly resisted a 
central gravitation tendency. 

119. We will assume as an hypothesis that our original 
solar nebula possessed the rotation period of Neptune of about 
165 years. Then, assuming this nebula of spheroidal form 
every section of which was condensing by gravitation through 
loss of heat towards the sun, with directive momentum in 
proportion to the original virtual tangential velocity of its 
parts, the mass of the nebula being for the present problem 
taken to be in density inversely as the squares of the distance 
of its^ parts, then, if the nebula could be entirely condensed 
upon the sun to its present state without loss by friction more 
than the excess of momentum due to gravitation in falling 
sunwards, the sun should after such entire condensation 
possess a peripheral velocity equal to the peripheral velocity 
of the original nebula. That is, in the case we assume, the 
velocity of the planet Neptune in its orbit. This may be 

120. The orbit of Neptune has a circumference of about 
17,253,000,000 miles and his revolution period is about 
60,000 days, equal to an absolute diurnal velocity of about 
287,000 miles in our sidereal day. The sun has a periphery 
of about 2,679,400 miles. Therefore, if the whole nebula 
condensed upon the sun without friction in shell-layers within 
the orbit of Neptune, the sun's linear velocity would be at 
its equator equal to the velocity of Neptune in its orbit, 
which would make its period of rotation somewhat under 
9^ days. Observation shows the sun to rotate in about 25 
days, so that the present diurnal velocity of the equatorial 
surface of the sun is 107,180 miles, that is only a little over 
2*6 of that which would be due to condensation of the 
enclosed nebula within the orbit of Neptune, upon the con- 
ditions proposed above, taken as a trial hypothesis. 


121. Upon the nebular theory the formation of any planet, 
say Neptune, could not have occurred until a large volume of 
exterior matter had condensed to this position. Indeed, the 
whole condensation must have taken place at the exterior or 
radiation surface of the nebula at a distance within which 
alone the planet could be formed, as exchanges of heat would 
prevent it condensing sunward, as before stated. Therefore 
the rarer portions of the nebula must have extended at an 
early period to a great distance beyond the orbit of Neptune, 
and the same extent of nebula must have been instrumental 
in sun-formation, although the condensation to form the sun 
might occur centrally from loss of general elasticity. If we 
assume the original radius of the nebula to be represented by 

= , where Y is the orbital velocity of Neptune and v 

that of the periphery of the sun, and D the distance of 
Neptune, we have for the distance d of an exterior planet 
in the present case for particles in gravitation-equilibrium 
according to the law of orbit about 38,000 millions of miles, 
that is if the nebulous matter was moving with equal angular 
velocity to that of the sun's present equatorial surface. 

122. This being the radius of equilibrium of a particle 
moving in a circular orbit at the distance given, would 
indicate upon the principles of gaseous condensation the 
fullest possible extent of our original nebula that could have 
been active in motive factors upon the sun, if its density 
diminished inversely as the square of its distance and the 
condensation was frictionless. In this theoretical calculation, 
therefore, we may take it that the angular motion of the 
nebula produced the angular motion of the sun, thus leaving 
direct gravitation in condensation to represent his heat. 

123. In the above eonstniction, although the motion of a 
planet may be used as the index of the extent of the original 
nebula at an early period, yet the planet's own formation may 
be excluded from the consideration of sun-formation, seeing 


that the entire mass of the planetary matter may not exceed 
T ^ part of the mass of the sun and that the result of its 
formation may be entirely different, the nebulous pressure 
on the sun by gravitation through loss of heat by radiation 
producing an intensely heated centre by concentration of 
exterior energy, as shown by Helmholtz ( 10), whereas the 
energy of the planet system is expressed more particularly in 
motive factors. So that gravitation under certain conditions 
may increase the orbital velocity of a planet, whereas it may 
be active in heat factors on the sun. 

124. The Momentum of a Planet. If we assume a zone- 
ring of nebula formed of uniformly distributed matter or 
otherwise to be detached from the sun's nebula, and to be 
moving at the orbital velocity which would be necessary for 
its detachment, and that all parts of the zone-ring are moving 
inter se at equal angular velocity according to the theory of 
Laplace ; then the entire momentum of the outer parts of the 
nebular zone, in falling towards the inner parts by conden- 
sation to form the planet, will carry with them the plus 
momentum of the outer parts which must appear in motive 
factors upon the planet when formed from condensation of 
such a system. If the angular velocity of the zone-ring 
at its outer parts were correlative with the orbital velocity of 
the inner parts, the gaseous system of the nebula would 
extend and not condense. It therefore becomes apparent, 
that to maintain the momentum of the original angular 
velocity of a nebular zone, the zone must remain in some 
way attached to the solar nebula during a large part of its 
condensation, and this momentum must be conserved by 
the intermotion of its parts. 

To consider the value of the motive factors of the parts of 
a condensing zone, we may take it from the period when its 
outer angular velocity equalled the orbital velocity of the 
next outer planet, that is at the period when its system was 
detached from the solar nebula. Upon this construction we 


may divide the entire momentum of a nebular planet-zone 
moving under the sun's attraction into two factors of motion 
angular motion and gravity. These we can distribute into 
two constants as the final condition of condensation : 
1. Permanent acceleration of rotation of a condensation 
within the zone, that is the future planet. 2. Acceleration of 
revolution to give the planet orbital motion. We may take it 
as an hypothesis that the sunward acceleration by gravity 
into the original momentum of the condensations of matter 
falling from the outer part of the system gives sufficient 
acceleration of revolution to establish the orbital motion : 
and that the difference between the angular velocities of the 
outer and inner parts of the zone-ring gives the rotational 
velocity to the planet as the probable action of the motive 
factors evident in the system proposed. 

125. The Orbital Velocity of a Planet derived from tJie 
centralizing gravity of the outer parts of a nebular zone falling 
upon the inner parts. Assume a planet newly forming by 
detachment of a zone at the periphery of the solar nebula, and 
that this zone is condensing most rapidly at the greatest 
distance from the sun. At this instant the periphery of the 
solar nebula must be moving at slightly less velocity than 
the detached zone, as before stated ; so that we make the 
velocity of the outer periphery of the nebula of the next 
inner planet zone-ring that of the orbital velocity of the next 
outer planet. We have no data for this in the case of 
Neptune, which must be referred to the extent of the sun's 
nebula at its peripheral velocity, but we may take any other 
two nearer planets whose mass is sufficient to allow of their 
being regarded as nebular formations, say Saturn and Jupiter. 
In these the linear velocity of the sun's nebula at its equator 
could not have been so high as the velocity of Saturn at the 
time the Saturn zone-ring was just detached from the sun's 
nebula. If we consider these orbital velocities, we find 
Saturn equals 510,452 miles diurnal velocity, that of the next 


inner planet Jupiter 689,855 miles ; therefore, to define the 
velocity of Jupiter in its orbit upon the data suggested we 
require plus 177,403 miles. The plus velocity is herein 
assumed to be derived from gravity of the matter falling 
towards the sun, that is entirely condensing towards the 
position of Jupiter. 

126. The simplest possible construction to show this is to 
assume that matter condensing from the outward part of the 
zone was moving just below the velocity of an inner part. 
Then this matter would not have sufficient momentum to 
maintain its tangential position, so that it must fall into 
an elliptical orbit of which its original position in the zone 
was its aphelion. If we assume that this matter could move 
without resistance, its velocity would increase according to 
the law of radii vectores until it reached its perihelion. And 
if we suppose this matter to form a planet at its perihelion 
position by encountering just sufficient resistance in surround- 
ing matter to retain its perihelion-radius by deflecting it 
from an elliptic to a circular orbit, this would represent the 
velocity of the inner planet moving according to the law 
of orbit. 

We cannot of course presume that this is the real condition, 
although it may represent its motive factors. The nebula 
being gaseous, would resist the direct continuity of the orbit 
of a condensed outer particle or aggregation of such particles, 
so that they could only fall sunward in spiral paths, possibly 
as flocculi ; and the whole system might possibly be more 
nearly represented as a pressure system upon the new-forming 
planet accelerating its motion than as a free motive one, but 
the dynamic effects would be the same. 

127. The Rotation of Planets. Assume a planet ring to be 
detached from the sun's nebula, and that henceforth this is a 
free elastic body revolving with tangential velocity in equi- 
librium with the attraction of gravity upon it to maintain 
its orbital position according to Kepler's third law. If the 



rotation of the peripheral band from any cause were slower 
than this, it could not leave the sun's surface. If it were 
faster its matter would fly off into space ; or, putting the 
matter practically, the zone-ring would contract or expand 
to its position according to the law of orbit. As regards the 
separate parts of the planet ring, it is assumed that these 
being at first a part of the sun's nebula in a highly gaseous 
condition would revolve in all parts in relatively static 
positions, exactly as though the ring were a solid body 
attached to the solar nebula. This is inferred from the fact 
that any intermotion of its interior parts in a gaseous body 
would be more frictional than that of equal angular velocity 
in which there would be no internal displacements. 

128. If therefore we assume that the planet ring revolved 
at first with equal angular velocity in all parts in mean 
gravitation-equilibrium for its distance from the sun centre, 
and that a condensation occurred in any part of the ring from 
pre-existence of a denser part which may have been caused 
by the intrusion of a comet, a shower of meteorites or other- 
wise, then all parts of the ring-system sufficiently near 
together to support a system of cohesion in gravitation- 
continuity would be drawn towards the denser part with 
velocity of approach inversely proportional to the square of 
the distance from the gravitation-centre less the resistance by 
friction within the system. In this case the tangential 
velocity of the ring being assumed to be at its outer surface 
in equation with gravitation for a circular orbit, there would 
be no tendency for the ring to leave its orbit, but matter 
would be continuously drawn towards the centre of attraction 
of the new-forming planet, in which case accommodation 
must be found for the volume of matter set free from its 
outer zonal position and attracted towards the new-forming 

129. Under the conditions proposed, we should have 
towards the planet's nucleus currents from different parts of 


the ring-system which would resemble in a certain manner 
similar currents in the atmosphere which find accommodation 
in cyclonic action in revolution about the centre of inertia of 
the motive system, as originally suggested by Descartes. In 
this manner the cohesion of the nebular ring would, in 
condensation under the circumscribing cyclonic action, tend 
to produce a rotatory nebular globe conserving a large part 
of the momentum of the original uniform angular motion it 
possessed in the former nebular ring when it was attached to 
the sun, and all the active gravitation forces brought about 
by its condensation towards the sun, deflected as they must 
necessarily be for accommodation of space within the new- 
forming planet's nebula. 

130. If we assume that all parts of the planet ring con- 
served the momentum due to the equal angular velocity at 
the period the ring left the sun's nebula which we are 
bound to do, or to account for the dissipation of this energy, 
the rotation of the planet finally formed ought to be 
consistent with the linear velocities of its parts. Under these 
conditions the periphery of the planet when formed should 
rotate with linear velocity equal to the excess of linear 
velocity of the outer part of the nebular ring over that of 
its inner parts. This proposition may be discussed by the 
aid of a diagram. 

131. Let a a" a!" (fig. 13) be a part of the outer circum- 
ference of the nebulous ring, b b lr b 111 the inner circumference 
towards the sun, and assume these circumferences to be 
bounding-planes of matter contracting towards the planet in 
the direction represented by a" to a' ', b" to b f . Let a" to b" 
be any small arc section of the nebular ring. Then the linear 
velocity of matter falling to a! being impressed upon a" 
assuming a' to a" moved with angular motion equal to that 
of the sun's centre, and a" to be condensed upon a' would 
set the planet in rotation with velocity proportional to the 
excess of the momentum of the matter condensed into the 



inertia of the portion of the planet formed at the time. The 
condensation of matter from # x/ , when the linear velocity was 
less than the angular velocity of the planet at a' b 1 in relation 
to the sun, would impress its momentum at b' in direction 
opposite to that of a' ', so that this would proportionally rotate 
the planet in the same direction as the outer matter from a!' 

Fig. 13. 

to a'. Now if all parts of the ring by accommodation con- 
served the momentum in condensation due to difference of 
length of arc and that due to attraction of the parts of the 
system upon themselves, by w r hich the axis of the planet 
would be set in rotation by cyclonic action, the final velocity 
of rotation of the planet would be approximately such that its 
periphery would possess a velocity equal to the differences of 
the velocities of the inner and outer parts of the nebular ring 
a a 11 a'" and b b" b' 11 , between which it was formed, upon the 
conditions of this proposition. 

132. It is assumed that a planet upon final condensation 
from a nebular ring will be placed very near the interior of 
the ring, since radiation into space, as before stated, can only 
take place from the outer parts, and not towards the nebulous 
sun, which at an early stage of the planet's formation would 
be heated sufficiently to maintain a nebulous state and be 
nearly as large in diameter as the inner surface of the ring. 
The nebular matter being continuous in the ring, will 
maintain cohesion or separate by flocculation in attraction 
towards the planet, as before stated, during its formation; 


so that the exterior condensation may be represented as 
showers of rain-drops falling upon the planet, carrying with 
them the tangential impulse due to their original motion and 
former position. This would occur so long as the interior 
heat of the system could maintain a purely nebular state. If 
the whole system was falling to the critical temperature of 
liquefaction of a large part of the nebula ( 94), or the for- 
mation was influenced by a ring of exterior heated matter, as 
possibly was the case with planets interior to Jupiter, the 
regularity of condensation proposed above would be materially 

133. To maintain the condensation continuously in a 
nebulous state from the zone-ring to the perfect planet, we 
must evidently possess a large volume of nebulous matter in 
this ring, as any thinly distributed system would condense 
quickly by radiation of its heat after separation from the sun 
into discrete matter, before a nebulous planet could be 
formed. Therefore it becomes probable that only the large 
planets of our system, Jupiter and Saturn, could condense 
under purely nebular conditions wherein the equal angular 
velocities of the system might be maintained. Further 
inference that these planets are of purely nebulous formation 
is found in part from their low specific gravity, due probably 
to the conservation of heat in their interiors, resembling in 
a certain degree the conditions of condensation of the sun. 

134. Rotation of Jupiter and Saturn. Taking the planets 
Saturn and Jupiter under the purely nebulous conditions 
proposed, let r and r be respectively the outer and inner 
radii or distances from the sun of two consecutive planets ; 
then rri will be the diameter of the nebular ring which we 
may assume as a gravitation system of circular cross-section, 
as shown fig. 9, p p' (p. 67), from which the inner planet will 
be formed, and 27r(r TI) the difference between the circum- 
ferences of the outer and inner parts of the ring. Let P be 
the diameter of the planet, and ?rP therefore its circumference, 



and H the number of sidereal hours in one complete revolu- 
tion round the sun. The peripheral space that the equatorial 
surface of the planet will describe for its orbit in time will be 


2 T r, and the theoretical term H 1? according to this 

proposition, will be 


= H 1 . The following table 

constructed upon these data will give the theoretical rotation- 
time in hours proposed approximately for a planet of equal 
density throughout formed from gaseous matter moving with 
equal angular velocity, upon this hypothesis. The T column 
is the observed time in hours as far as known. 

135. It is seen that the virtual velocities of the parts of 
the condensation would depend upon their final position in 
relation to the centre of the planet. The nearer the centre 
the greater the angular motion impressed upon the planet 
from an equal linear velocity, so that a planet dense at its 
centre, and the reverse at its periphery, would upon final 
condensation have greater rotation-velocity than one due to 
the condensation of equally dense consecutive shell-layers 
over its surface to form its mass. In the following table the 
planets are taken as being of uniform density throughout. 

Table of Theoretical Rotation of Planets formed under 
purely Nebular Conditions* 


Difference of 
circumference of 
ring 271-0--^). 

of planet 
X hours =7rPH. 

rotation Hj. 

rotation T. 

Jupiter ... 





Saturn . . . 





Uranus ... 




136. In the above table the acceleration of rotation or of 


orbit due to the gravitation of exterior matter falling to a 
nearer position to the sun is not calculated, as this may be 
taken up in acceleration of the planet's orbital motion, as 
before proposed. In the last column Jupiter appears of 
greater rotation-velocity than the theory demands unless it 
increases in density towards its centre ; it would therefore 
appear, if the hypothesis given be strictly true as regards the 
only factor of motion, that the density of this planet must 
increase greatly from its centre to its measured perimeter. 
According to this hypothesis, to give a velocity of rotation 
of 9*868 hours for a planet of equal density throughout, its 
diameter should be only 75*248 miles. If it is partly solid 
and partly gaseous this number may express the diameter at 
mean density. It may be its diameter in four millions of 
years hence. Further the whole system of the planets may 
have been derived from nebular zones which at a period of 
early formation extended much beyond their present orbits, 
so that the planetary system may have contracted throughout 
its entire extent. This contraction would .give excess of 
rotation beyond that resulting from the principles discussed 
above, even assuming a part of the energy may have been 
lost in the friction of planetary formation. 

137. Although all difficulties with respect to acceleration 
of Jupiter's rotation upon the theory proposed may be removed 
by the suggestions given above, at the same time there is 
no doubt that the observations of the surface of this planet 
are very perplexing. The surface appears to be in violent 
agitation. It is not entirely improbable that this planet, 
quite recently as measured by astronomical past time, 
possessed a ring-system similar to that of Saturn. This ring, 
through some disturbing cause, the intrusion of a comet or 
other matter, was probably thrown out of gravitation-equili- 
brium, and the ring-matter, having stronger attraction to 
the body of the planet than to the parts of the ring so as to 
draw them together to form a satellite or satellites, the 


separate parts of the ring are now drifting in concentric 
orbits upon and around the planet with greater velocity than 
that of its surface. In this case the heat engendered at the 
surface of the planet by condensations or collisions of meteoric 
matter projected thereon would produce a nebulous atmo- 
sphere permitting only the approach of the exterior meteoric- 
ring matter, which is probably in the form of dust, to drift 
under the resistance it encounters in spiral or cyclonic paths 
about the planet's equator and outward from it. This may 
produce the present surface appearance, which may be similar 
in motion to cyclonic areas in the atmosphere about our 
equator. The virtual velocity of the parts of the broken ring- 
being greater than that of the surface of the planet, this 
matter in drifting over and covering the surface of the planet 
would present the only measurable part open to our observa- 
tion, the planet itself being entirely obscured. 

138. In Jupiter we have not the low density which we 
have in Saturn to indicate formation under purely nebular 
conditions, so that the reasons given in the two previous 
paragraphs are quite sufficient to account for the excess of 
velocity over that required by the theory of our table. If 
necessary, it would not be difficult to find others. If the 
zone-ring was originally elliptical, and the planet condensed 
at its perihelion position, the difference of linear ratios of the 
outer and inner surfaces (fig. 13, a a, b b) would fully account 
for the excess of rotation. Again, it is not impossible that 
exterior matter moving at greater velocity drifted into the 
planet during formation. 

139. In the planet Saturn we have evidence of purely 
nebulous conditions in formation, in which the surrounding 
equilibrium of matter was so perfect that central rings of 
condensed free matter were possible of formation round its 
equator. In this case we have a rotation period in nearly 
exact equation with the tangential velocities of the part of 
the exterior nebula from which the planet was formed upon 


the theory proposed. Nevertheless it must be in some degree 
accidental that this comes so nearly into agreement with my 
theory in this planet. It is in all probability much more 
dense in its central parts than at its periphery ; it should 
possess, therefore, greater velocity than it does, but this 
difference may well represent the friction of the system in 

140. Further, with regard to Jupiter and Saturn, it is 
quite uncertain whether we make sufficient allowance for 

in measuring such bright bodies as these planets. 
The mean measurements of Venus as a bright body by 
Hartwig, Kaiser, Airy, and Ambronn give H /f '593 at a 
distance equal to its mean transit-position. The measurement 
of Venus by Auwers as a dark body in transit gives 16^'SOl, 
from which we may conclude the true measurement of Venus 
is probably 17 // *197 ; applying a similar reduction to Jupiter, 
which it is impossible to measure in the same manner, we 
have its diameter about 80,000 miles. This proportion is also 
confirmed by some measurements made by myself of the 
iridescence of bright bodies under the microscope. It is 
evident also in the apparent thickness of the filament of an 
incandescent electric light, that may be compared with its 
reflection in a polished surface of black glass, which suggests 
a possible mode of measurement of Jupiter and Venus by 
reflection from a black surface. 

141. Uranus is added to the table, which gives, upon the 
theory of the rotation of Jupiter and Saturn, a period of 
12*34 hours. This planet, however, if we take the direction 
of its satellites as an index, is moving probably, but not 
necessarily, in the reverse direction, which might occur from 
its formation from discrete particles, according to the theory 
of Faye *. This theory may be shown by the same diagram. 
In this case matter is assumed to be moving in free orbits 

* Faye, < Sur 1'Origine du Monde,' 2nd edit. p. 117. 


round the sun, and the particles along the arc (fig. 13, a, a! a ff ) 
assumed to have less virtual velocity than those along b b' b", 
consequently the planet would move in the reverse direction. 
Another plan of rotation will, however, be suggested further 

142. Rotations of the Asteroids. The rotations of planets 
inferior to Jupiter would, of course, depend upon the mode 
of their formation. The attenuated plane of the nebula pro- 
posed for the formation of the Asteroids would leave these 
condensations about this plane so as to form an early break 
in the nebular system of the sun, and the henceforth separate 
nebular system of Jupiter ; so long as the Asteroids remained 
in the orbits of their original formation, they would take 
rotation periods consistent with the principles proposed for 
the formation of purely nebulous planets considered above, 
little affected by the decentralizing action of Jupiter or its 
original ring-system. Any of these minor planets, if formed 
through separate condensations of concretions of smaller 
planets, or of meteoric matter by after collisions, in the 
crossings of orbits, would have their original rotation due to 
condensation under nebular conditions diminished by the loss 
of the directive-momentum of rotation, which would be 
converted into heat at the time of collision. The absolute 
condition of these bodies must remain altogether speculative, 
our present limited knowledge being insufficient to obtain 
data for the actual rotation-periods necessary for its con- 
sideration. They most probably move with different rotation- 
velocities, and it is not improbable that some rotate in the 
reverse direction. 

143. Mars. If Mars was formed directly under the 
nebular conditions proposed for the superior planets, from a 
nebular ring which extended to the near asteroid ^Ethra, it 
would possess a rotation only slightly in excess of its revolu- 
tion-period. If we assumed a wider area for the ring, possibly 
to include the formation of 2Ethra in another or opposite part 


of the same zone-ring, we might theoretically give the 
rotation-period of Mars its actual or any value we please up 
to or beyond that of Jupiter. It is, however, probable that 
other than purely nebular conditions influenced the formation 
of this planet. Mars, the Earth, and the inferior planets 
possess a density-system which very probably indicates con- 
tacts or collisions of solid matter to form the interior parts. 
Therefore the nebular conditions proposed for the superior 
planets of low density do not hold entirely for index of 
rotation-period in this case for planets of high density, as it 
would be considerably modified by the discrete mode of 

144. Laplace suggested that a nebular zone-ring would 
break up into many parts, which would draw together and 
form separate nebular globes. These being formed at slightly 
varying distances from the sun, would finally coalesce and 
form a single nebular globe. In so attenuated a system as 
the zone of Mars would be upon this theory, such nebular 
globes as may have been produced by early condensations 
would probably be condensed by radiation, before union could 
be effected into a single globe. Therefore the separate globes 
or units of condensation might at a certain period be repre- 
sented by a band of planetoids formed by condensations in 
different parts of the original nebulous band, these planetoids 
being formed within the band at only slightly varying dis- 
tances from the sun. If the orbits of the separate globes were 
elliptical and of nearly coincident period, they would con- 
stantly approach in opposition, come into collision, and by 
cementation, in which a certain amount of heat would be 
developed, produce finally after cooling the diversity of 
surface we observe upon this planet. It is possible that this 
form of collision may again occur at a definite calculable 
period with the near asteroid .ZEthra. 

145. The rotation-period of Mars, as of other planets, must 
necessarily represent the sum of the directive momenta of 


the factors of matter which formed the planet, and this might 
be fixed upon purely nebular conditions as with the planets 
Jupiter and Saturn, allowing time for the nebula to con- 
tract upon the sun only, without planet-formation as before 
proposed. The general conditions of density of the surface- 
formation of Mars and the presence of near Asteroids do not 
indicate equable nebular conditions at its formation. This 
does not, however, infer that it was not at one period partly a 
nebulous globe, the gravity of which was superior to that of 
any near body of condensed matter ; indeed, this is probable 
for the formation of satellites upon principles to be discussed 
further on. 

146. Rotation of the Earth. If we take the same purely 
nebular conditions as proposed for the formation of the 
large planets by the condensation of matter carrying virtual 
velocities of the parts of the ring-system to the planet, and 
assume under like conditions that the earth's nebula ex- 
tended at one period to the orbit of Mars, we shall arrive at 
a rotational velocity much greater than that which we observe. 
But in the discoidal nebular system proposed, fig. 10, p. 85, it 
would be extremely improbable that the actual earth-forming 
nebula in contact with it ever possessed this radius. By 
the principles of sun-condensation without planet-formation 
already discussed, and its condensation when matter just 
interior to its orbit was at a critical temperature, we may fix 
the earth's nebular ring-orbit consistent with its period of 

rotation, as before, by the formula -p-a- , considering the 

density of the earth and a certain extent of space between its 
orbit and that of Mars for its volume. The probability is 
that the greater part of the Earth-Mars interspace was 
condensed at an early period into separate smaller nebular 
planets, as just proposed for Mars, and showers of meteorites 
at its outer parts were moving in very elliptic orbits. The 
condensation which may represent the nucleus nebula of the 


earth-moon was possibly at about its present mean radial 
distance from the sun. Under these conditions the iouter 
condensations, whether nebulous, liquid, or solid, assumed of 
eccentric orbits, would be drawn at perihelion towards and 
into the earth's nebula, if these discrete condensations at the 
period extended, as was probably the case, beyond the radius 
of the moon's orbit. Further, in an attenuated ring-system, 
as the earth's system may have been at an early period, while 
rapidly cooling, condensations would occur at different parts 
of the ring, and at slightly varying distances from the sun, 
as before suggested, so that ultimate collision must occur 
between them and the earth. Under such conditions we 
should have through condensation variety of density and 
surface-conformity, and rotation in composition with all the 
motive factors of the earth's formation. The development of 
heat under the process of formation may have maintained a 
considerable nebula round the earth at an early period, but 
not being an entirely nebular formation its motion of rotation 
would be much slower than that calculated for Jupiter and 
Saturn if taken under purely nebular conditions. This 
matter will be reconsidered in detail further on. 

147. The rotations of Venus and Mercury would be subject 
to the same conditions as that of the Earth and Mars. How 
far these planets may be formed from nebular or from discrete 
matter it is impossible to say. If formed from nebular 
matter, the motion would be greater than that of the Earth 
actually. If formed from discrete matter, produced by the 
general lowering of the temperature of the surrounding 
nebula to its critical temperature, it would be less ; or these 
factors might act conjointly, so that the rotation might 
be nil. 



148. Revolution of Satellites with Direct Motion. In the 
revolution of satellites around their primaries under purely 
nebular conditions, no other law could hold for the conden- 
sation of nebular matter than that which must hold for 
condensation upon the primary ; so that, if the motion of the 
planet's perimeter were equal to the difference between the 
linear velocities of the inner and outer parts of the primitive 
nebular ring from which it was condensed, the satellite's 
revolution should be consistent with this under the same 
mode of formation. Thus, taking the satellite's distance and 
revolution period, the virtual velocity of the satellite should 
equal the difference of rates of the extreme parts of the ring, 
previously expressed as 2?r(r r^). At the same time this 
must be taken as representing the angular motion of the 
entire exterior nebular system within which the satellite or 
satellites, or any ring-system, as in the case of Saturn, 
were formed. As soon as the satellites were formed, 
the condition given, 127, for planet formation must hold 
in the condensation of an attenuated system about the 
planet. The motion and position of the satellites, if there are 
more than one, must finally rest in relation to the planet 
according to Kepler's third law, the squares of the numbers 


representing their periodic times varying as the cubes of the 
numbers representing their mean distances, subject to such 
influences as the sun's attraction may produce upon the 
system. Under these conditions, the energy of the equal 
angular velocity of the original nebular ring may be split up 
into factors represented by the separate satellites and the 
general equality of the original angular velocity be lost, 
while retaining the mean momentum of the original sur- 
rounding nebular system about the planet. So that, in a 
purely nebular system not previously locally condensed, we 
ought to find the momentum of the angular velocity of the 
primitive nebula fairly represented in its entirety in the 
central system of the planet and all its satellites. Further, 
under these conditions of nebular or gaseous formation, the 
motion of the satellites must be direct with respect to the 
planet. Therefore it cannot include the systems of Uranus 
and Neptune, wherein the motions of the satellites are retro- 
grade. This leaves us for consideration only the outer 
planets, herein presumed to be formed under purely nebular 
conditions, which have direct motion, or, more particularly, 
the satellites of Jupiter and Saturn. These satellites, as 
before stated, under the conditions given, cannot be taken 
individually, but as a mean of mass and motion of the whole 
contained in one system. 

149. Revolution of the Satellites of Jupiter and Saturn. 
Under the conditions given, let 2?r(r rj represent the 
mean momentum of the nebular ring active upon condensa- 
tion in forming the planet and its satellites. Let 

where S represents the diameter of the orbit of the satellite, 
H the hours in the sidereal year of the planet, and Hj the 
hours in one complete revolution of the satellite round its 
planet. In this manner, the revolution of the satellites of 



any system in the aggregate should agree with the angular 
rate of motion of the equator of the planet, and this with the 
momentum of the nebular atmosphere surrounding the planet 
from which the whole system of satellites is assumed to be 
formed. As experiment, we may take each of the satellites 
of Saturn and of Jupiter to be formed of equal quantities of 
nebular matter at about their present positions separately, and 
divide by their number so as to find the mean place of the 
imaginary satellite we desire to consider as the unit of the 
system. The result of this is shown in the following table : 

Table of mean Rotational Velocities of Satellites. 

Mean position of 



2*r(r-r 1 ). 


Saturn's satellites and ring 
Jupiter's satellites . . 


2,490 950 000 

- 275,800,000 
+286 500 000 

150. We see again in this an excess of velocity in the 
satellites of Jupiter which is consistent with the excess of 
motion over that derived from 27r(V r { ) observed in the 
planet. This may infer an intrusion within the nebula 
previous to the formation of the planet of a large mass of 
matter moving with greater velocity, but so as to produce the 
same direction of rotation as that of the original nebular ring, 
as before proposed, which is again consistent with Jupiter's 
greater motion and abnormal mass. The difference shown, 
however, is made much less by taking the mean motion of 
the satellites in conjunction with their masses, the outer 
satellites of Jupiter being of fgreater mass than the inner 
ones. And in the same manner the mass of Saturn's rings 
exceeds that of its satellites, so that this difference would be 


again diminished, and would come as fairly to my 


theory as the conditions probably would admit of calculation 
in a system wherein there may have been intrusion of exterior 
matter, cometary or other, during the formation. 

151. Satellites of Mars. These small bodies move at a 
higher velocity than the equatorial surface of the planet. It 
is therefore clear that they could not have formed a part of 
a nebular system moving at equal angular velocity with the 
planet, assuming the planet was entirely condensed from the 
nebula and moved originally only at its present rotational 
period. By the condensation of a nebular zone as defined 
above for the satellites of Jupiter and Saturn, the rotation 
periods of the satellites of Mars would be made consistent 
by treating them as revolution systems, 127. It is clear, 
however, as before suggested, that we have not in Mars a 
nebulous condensation of the kind that we have in the outer 
planets, where the condensation has produced a mass of 
small specific density. The probability is that the nebula of 
Mars possessed at a certain period a rotation consistent with 
the revolution of its satellites, but that the planet was of 
smaller mass, the whole system appearing as a planetary 
nebula. That about this period the planet entered into col- 
lision with one or more planetoids of smaller mass than itself, 
which were previously condensed to solid form. These plane- 
toids may have penetrated its nebulous atmosphere without 
materially changing its rotational velocity, but have reduced 
that of the planet upon collision. Such collisions would 
develop great heat, partially liquefying the solid planetoid, 
and produce by cementation with it a partial protrusion 
of matter beyond the sphere, in which we have an index of 
the surface configuration. In this manner, the momentum 
of the outer parts of the nebula would be maintained, although 
the penetration by a planetoid to the centre would possibly 
cause sufficient disturbance of equilibrium in the nebula for 
the satellite-zone to immediately commence the formation of 
the satellite. 


152. The orbit-position of a satellite can only follow 
Kepler's third law. In a satellite-zone open to a system of 
condensation at the outer radiation-surface of its nebula 
moving at equal angular velocity, in which refractory matter 
of the outer nebula condenses first, the condensed units will 
move through the resistance of residual gas under the influ- 
ence of gravity in spiral lines, until they reach the central 
condensation, unless an orbital velocity position is found 
according to Newton's Law * ; and in this position the 
satellite must be formed. Therefore a satellite may be formed 
at any distance from its planet. If matter falling upon a 
planet possesses less tangential momentum than that which 
produces orbital velocity at any position above the planet's 
surface, satellite formation would be impossible as such matter 
would fall to the body of the planet. 

153. The Moon possesses a higher virtual velocity than the 
equatorial surface of the earth in the proportion of 2288*43, 
the moon's mean hourly motion, to 1037'6, the hourly motion 
of the earth's equator. Assuming the earth system to have 
been entirely nebulous, and the nebulous matter to condense 
in equal shell-layers over the earth, commencing from the 
time of the condensation of the moon from gaseous matter, 
decreasing in density inversely as the square of the distance 
from the earth's centre ; then the earth should possess nearly 
the same initial velocity at its equator as that of the moon in 
its orbit, assuming no friction of formation developed into 
heat upon the earth's surface at the time. In many ways 
this is improbable, as the whole system does not confirm the 
earth's formation under entirely nebular conditions. It is 
therefore probable that the earth's nucleus was a considerable 
liquid, or partially solid, mass before the time of the moon's 
early condensation. Further, the configuration of the earth's 
surface indicates that it was probably formed in part by 

* ' Principia,' Lib. ii., prop. xv. 


concretion of planetoids or meteoric matter projected from 
exterior space into its nebular ring, which retarded its 
velocity and disturbed its axis of rotation. Either of the 
above-stated conditions, independently of the friction of 
formation or of tidal friction, if this may be included, would 
fully account for the earth's minus rotational velocity com- 
pared with the linear velocity of the moon, which must 
necessarily follow the law of orbit. At the same time it is 
highly probable that the nebular system of the earth extended 
much beyond the orbit of the moon at the time of the com- 
mencement of the moon's condensation *. 

154. The rotation of the moon taking place in exactly the 
same time as that of its revolution round the earth, infers 
that the condensation of the moon was at first into a complete 
narrow ring, probably liquid at one period, therefore moving 
in all parts at equal angular velocity to the earth. If it 
condensed as a globe from a wide vapourous ring moving 
in all parts at equal angular velocity, as previously proposed 
for the outer planets, it would then possess a direct motion in 
rotation in excess of its revolution period. If it was formed 
from local condensation at first into discrete nebular matter, 
it would possess a reverse rotation f ; combination of these 
factors, direct and reverse, might give its present rotation, 
but the proposition of the early formation being a narrow 
ring appears to me more probable to bring about the exactly 
equal periods of rotation and revolution. 

155. The condition of condensation in rings which after- 
wards formed satellites appears to be difficult of conception 
to some astronomers. This must be so, if the density of all 
parts of the ring is assumed to be uniform, as it appears to 
be approximately in the rings of Saturn. But this system of 
equilibrium appears to be exceptional; if the rings move at 

* See the Author's paper, Brit. Assoc. Reports, 1885, p. 915. 
t See Faye's Theorem, ' Sur TOrigine du Monde/ p. 117. 



orbital velocity, the equilibrium of its matter from tangential 
momentum and gravity is such that, irrespectively of its 
rotation, its matter may be considered to float on a frictionless 
plane ; so that the initial gravity of its mass acts directly 
upon itself. Therefore, if the nebular moon-zone were 
unequally distributed, its condensation would be direct to 
the densest part. 

Suppose the condensed zone represented by fig. 14 a, then 
its point of tension would be at x. Assume this point to 
separate by the initial slow action of the entire gravity of the 
ring. Then its matter near the point of separation would 
gather into two semi-globular terminals, fig. 14 b, x'. Initial 

Fig. 14. 

gravity would now act in opposition to the direct momentum 
of the mass, accelerating the velocity of the limb b and re- 
tarding that of c. So that at some point about d the ring 
would condense into a globular mass, by its extreme conden- 
sations coming together. 

156. Retrograde Motion of Satellites. The conditions 
proposed for the direct motion of the satellites previously 
considered as derived from the condensation of the gaseous 
nebula, could not possibly hold upon the hypothesis given for 
the satellites of Uranus and Neptune, which move in the 
retrograde direction. It may appear, so far as the direc- 
tion only of this motion is concerned, that this would be 
demonstrated by the theory of M. Faye, 141, as due to their 


formation from discrete matter, the satellites being condensed 
from a ring of such matter, every particle of which was 
originally moving in a free orbit in gravitation equilibrium ; 
this theory may possibly be applicable to the satellites of 
Neptune. There are, however, many peculiarities about the 
satellites of Uranus which do not admit of this hypothesis. 
They move in orbits inclined nearly 80 to the planet's orbit- 
plane, so that the differences of velocity of the parts of the 
orbit of any assumed nebular ring or matter caused by the 
differences of distance of its parts from the sun would be 
very small. These conditions would, therefore, cause the 
satellites to revolve at a very slow rate, that is, at about 
2^0 f that observed. We must, therefore, certainly look 
for additional causes for which other suggestions may be 

157. If discrete matter was formed at the limits of the 
solar nebula, when this was moving at less than orbital 
velocity, as before supposed, and the early discrete matter 
was formed of the more refractory matter so as to leave the 
more attenuated, less refractory nebula to form a resistance 
to the centralizing condensation of the discrete matter, then 
this matter might carry the momentum of its former angular 
velocity to the inner, denser nebular matter which formed the 
sun at the time, leaving the residual matter slowly condensing 
at less angular velocity. Such a system would produce a solar 
nebula moving at higher velocity than its peripheral lighter 
outward parts. I have endeavoured to show in my work upon 
Fluids* that every fluid system in rotation, cyclonic or other, 
engenders in the surrounding fluid or medium which offers 
resistance to its direct motion an opposite direction of motion 
of rotation around its borders. By this action the central 
rotating fluid attains a kind of rolling contact upon the 

* * Experimental Researches into the Properties and Motions of Fluids, 
1881, p. 224 et seq. 


surrounding, more static fluid, which produces by this mode 
of rotation the least frictional resistance to the motion, in 
what I term friction-whirls to the direct flowing system. 
Upon this principle, if the central system in this case were 
condensing the more refractory pneuma into nebula with 
increase of angular velocity by the effects of gravity, then the 
surrounding, less refractory pneuma, less motive in the 
direction of the central system by the resistance of surrounding 
matter not moving at equal velocity or possibly in the same 
direction, would offer a certain amount of resistance at the 
periphery of the central condensation of nebulous matter. 
This is shown by the lateral form of motion to a current, fig. 5, 
p. 44 ; it may possibly be better explained by a diagram. 
Let one plane of the interstellar space, subject to the superior 

Fig. 15. 

attraction of our sun, be represented by the circumscribing 
outline N, fig. 15, along which line the sun's attraction is 
assumed to be in equation with that of other near stars. 
Let S be the centre or sun ; and let the central nebula be 
bounded by the circle 0, the arrow near the line showing the 
direction of rotation. Then in the space bounded by N on 
one side and upon the other the matter would be less 
motive than that of the central system ; and this would 


rotate U in the reverse direction, as shown by the arrow. 
This form of motion is shown by many experiments in my 
work on " Fluids " already quoted. 

158. If the central system contracted by the influence of 
gravitation and other conditions already discussed, say, to the 
diameter represented by jt>, which we may take at the orbit 
of Neptune, then the whirl system U would be also con- 
tracting upon itself; and being also attracted by gravitation, 
it would become a motive part of the solar system S, falling 
into the same direction of revolution but with the opposite 
direction of initial rotation, just as a roller or ball-bearing 
moving between the surfaces of an inner and outer cylinder 
progresses along the surface of the inner cylinder in its direc- 
tion of revolution, but at the same time takes the reverse 
direction of rotation. Upon this theory it may be inferred 
that Neptune and Uranus were possibly perfect rotatory 
systems of nebulous matter before they were incorporated 
into the central or solar nebular system. 

159. In the above construction, considering the inclinations 
of the orbits of the satellites of Uranus, we have at first the 
loss of directive momentum due to the differences of velocity 
of the parts of any ring-system that could have formed and 
rotated these satellites in the plane of the planet's orbit, 
so that the surrounding matter in the satellite's orbit-plane 
became more subject to resistance than if the planes of orbits 
of the planet and satellite were nearly parallel. The motion 
for reverse rotation under fluid resistance is active laterally, 
exactly as in the plane of motion, and resistances are more 
likely to occur exterior to the orbit-plane than in it. The 
form of nebula at a certain period of formation best adapted 
to produce the system of the satellites of Uranus, is possibly 
that of y I. 176 in Coma Berenices, where the nebula appears 
to turn up at one terminal at a considerable angle to its mean 
plane of rotation. 

160. Under the same principles as given in the above 


hypothesis, it may be suggested that the contact form of 
reverse rotation may have been produced in a nebular con- 
densation having either no rotation or a reverse rotation, 
being attracted into the sun's nebula. Such a condensation 
may have occurred in an interspatial position between the 
nearly equal attractions of several stars, as, for instance, 
in the position outwards between b and of or a and b 1 of 
fig. 11, p. 88. If a nebula so formed drifted by small excess 
of gravity towards our sun when it was in a nebular state, 
it would enter the frictional system of the borders of the 
attenuated solar nebula. The inertia of the newly introduced 
planetary nebula would resist the rotation of the solar nebula, 
which would therefore produce a motion of rolling contact in 
the meeting-plane between the two systems, as before sug- 
gested, as the least frictional form of fluid motion. In this 
manner a kind of wheel-and-pinion motion would be induced, 
in which the smaller planetary nebula would represent the 
pinion moving in a reverse direction to the solar nebula 




161. If we can accept the idea of a universal pneuma, 
23, and that this pneuma was motive in rotation, separating 
and condensing into separate systems, 43, a condition of 
original motion in matter that may not be unlike that pro- 
posed in the theory of Descartes and Faye * ; then all separate 
parts of the system must continue motive under condensation 
to take up the general momentum of the system. 

162. If we consider the relatively small volume that would 
be circumscribed by our solar nebular system, if taken to be 
in its original nebular state of spheroidal form at a period 
when it was circumscribed within the orbit of Neptune, as 
compared with the mean distance between our sun and the 
nearer surrounding stars, we find what a relatively small space 
the orbits of our planets occupy. So that if the original 
pneuma at its earliest period extended to the mean of inter- 
stellar space about our sun as suggested, we can only imagine 
that many millions of local condensations due to exterior 
radiation were formed in exterior parts of the system. These 
condensations would afterwards only slowly drift sunward, 
but they would still carry with them the same factors of 
original rotative influences and the influence of the attraction 
to near-surrounding matter, as before discussed, 70. 

163. Therefore, taking the original pneuma system as shown 

* <Sur POrigine du Monde,' 2nd edit. p. 101 et seq. 


in one imaginary plane, fig. 8, AB (p. 59), extending to the 
mean distance of our sun and a near star, we can but imagine 
that at an early period there may have been many millions of 
local rotatory systems of matter condensed to a nebular con- 
dition in a free state, which would be moving from the outer 
pneuma system from the positions a, 6, c, d sunward by central 
attraction. We may group such systems together as comets. 

164. In the above construction we may suppose that the 
cometary system was the earliest prevailing system of the 
outer condensation of matter ; that the comets which were 
formed by local condensation to nebulae were most generallv 
absorbed into the solar-planetary system when this was of 
immense volume ; so that at a certain period our solar system 
would have appeared, if viewed from a great distance, as an 
immense floccular system wherein the exterior cometary 
condensation would appear incandescent from friction and 
electrical excitation within the interior nebular condensation, 
and from intense chemical action in the condensation of the 
pneuma to nebula. These flocculi would be drifting sunward 
by the effects of their attraction and the small resistance of 
the surrounding attenuated pneuma in spiral paths, similar 
to the spiral nebulae 31 M, 81 M, 56 $, 168 $, &c., as 
before proposed. 

Another condition that must mark the formation of comets 
considered as exterior condensations of the pneuma system, 
is that tne direction of orbits must have been influenced by 
the gravitational effects of the larger mass condensation 
forming at the time upon the scheme proposed in 83 and 
illustrated by fig. 8, p. 59. Therefore, comets must have 
been formed in many cases in series, from attractions a, b, c, d 
of this figure, taking one orbit-plane, but with varying 
eccentricities of orbit, depending upon the amount of original 
tangential impulse each comet possessed upon starting sun- 
ward, as the conditions entail. 

165. Upon these principles the comets and cometary matter 


we possess at the present time in our ancient solar system 
represent only the waifs and strays formed by exterior con- 
densations which have escaped absorption, that have arrived 
from interstellar space after the sun had condensed to nearly 
its present volume, 68. The relatively small number that 
remain may also, through the disturbing influences of the 
antagonism of initial centralizations and solar attractions, 
hereafter experience internal strains and collisions within their 
systems about perihelion which may cause their disintegration 
or deformation, so that they may be lost to astronomical 
observation in the future, under conditions to be discussed. 

166. Although the earlier condensations of pneuma to 
nebula exterior to our motive solar-planetary system would 
produce, upon conditions discussed, rotatory systems of nebulae, 
we can scarcely imagine that such nebular conditions could be 
generally maintained in exterior matter until the present time 
under the excessive radiation of their original heat into space. 
Therefore, we have the extreme probability that cometary 
matter is at the present time largely condensed into solid small 
units of meteoric or planetary matter. 

167. The theory of association of comets with meteorites 
is as old as Anaxagoras, who states that comets are the con- 
gregation of wandering stars that approach so near to each 
other that they appear to touch *. This theory, for certain 
factors of cometary existence, is greatly strengthened by the 
discovery that our known meteor-streams follow in the orbit 
of certain comets, as pointed out by Schiaparelli for the 
August meteors, and for other meteor-streams by Oppolzer, 
0. F. "W. Peters, Prof. Newton, and others, which theory is 
also ably supported by Prof. Lockyer f. 

Under this theory the difficulty of finding cause for the 
self-illumination of such streams, if they form comets, is very 

* Stanley's < History of Philosophy,' p. 64. 

t Astr. Nach. No. 1384. Lockyer's Meteoritic Hypothesis/ p. 138. 


great. Professor Tait has endeavoured to prove that such 
meteorites would be subject to sufficient collisions among 
themselves to account for the light * ; but if the meteorites 
follow one another in streams, there must be very small 
differences of velocity between them, and if they approach 
one another, assuming them widely distributed miles apart, 
their collisions must be very gentle, and produce very little 
light, if any. Otherwise, the observations of Mr. Denning 
upon the Biela meteors show that they possess great diffuse- 
ness of radiation, so that their paths appear to diverge from 
an area rather than from a point of the sky, indicating inter- 
motion among their separate units. 

168. There is another objection to the meteoric-swarm 
theory which has not been attempted to be met, or even 
suggested that I am aware of, which is, that the separate 
units of the comet must, under the conditions of this theory, 
necessarily follow Kepler's third law of orbital motion, that the 
squares of the members representing the periodic times of the 
separate meteorites must vary as the cubes of their mean dis- 
tances from the sun. Therefore, assuming any comet to have 
a tail of one million miles in diameter, as commonly observed 
for the larger comets, the velocity of the outer meteorites of the 
swarm must be very much less than that of the inner ones 
nearer the sun. Under this condition, if the comet were a 
swarm and at a certain period from some unknown cause of 
the symmetrical form common to comets, as, for instance, 
that of Halley or of the great comet of Sept. 1882, or 
other, Plate III. A, i,j 9 as the parts of the swarm would be 
actuated by various velocities, it could retain this form for a 
short time only. So that on its return to perihelion, for 
instance, it could only be represented at most by a scattered 
band of meteorites spread over a great distance in space that 
could never again appear as a comet. To take a self-evident 

* Edin. E. Soc. Proc. 1879, p. 367. 


case of the conditions in question, assume our moon at opposi- 
tion to take a solar orbital motion without revolution around 
the earth, then it is clear by Kepler's third law that we should 
soon leave it behind us in space, so that it would become in 
time in conjunction with the opposite part of our orbit. 
Indeed the only condition possible for a meteoric theory in 
which a comet can retain a symmetrical form, is that the 
system of meteorites that form the comet should be in revolu- 
tion about the centre of inertia of the system, moving in 
elliptical orbits, in the same manner as the comet moves about 
the sun and as satellites are in revolution about their planets. 
This will be more particularly considered presently. 

169. Comets of long period. The general principles of 
direction of orbit from local condensations at a great distance 
from the sun have been discussed, 71. We have now, there- 
fore, only to consider the probable intermotion of the parts 
of such distant condensations as may possibly produce comets 
of the symmetrical form we observe in them upon planetary 
conditions, and, therefore, such as are outwardly, as it appears 
to me, evidently moving under the direction of symmetrical 
orbital law. 

In the extensive volume of pneuma considered as the 
extreme field of comet-formation, which would be subject to 
the influence of the near stars almost as much as that of our 
sun, 76, the centralizing influence of gravitation would 
have little effect in changing the natural formation of in- 
dividual systems of matter after they were once constituted. 
Therefore, assuming original motion in the pneuma such 
as we have found necessary for the formation of our solar- 
planetary system, 62, such motion must, as before stated, 
have extended to all parts of the solar pneuma. If any original 
isolated system, of large volume in its original state, were in 
slow revolution with its parts moving at equal angular velocity, 
then upon its condensation to a smaller volume its rotative 
velocity would increase, as previously discussed for solar 


rotation, 116. This rotation of any part of the system would 
be maintained in projection sunward, and if in free matter 
projected from a great distance, it would form a comet of 
long period. 

170. Comets of short period. These possibly depended in 
many instances for their orbit upon deflection of the matter 
of the comets of long period by the influence of planetary 
attractions. At any early epoch such long-period comets, 
moving at high velocities by accumulated gravitation in 
falling from a distant part of space, would fall into the solar- 
planetary nebula or into any detached zone-ring of nebular 
matter which may have been present at the time moving at 
orbital velocity round the sun. In such a case the motion 
of the comet would be retarded by the nebulous matter and 
enter into composition with the motion of part of the zone, 
or, if not incorporated with it, it would be deflected by it 
from its original orbit into a less eccentric orbit. Comets of 
short period would also be formed upon local disturbance at 
any part of a planet-forming zone-system by a local condensa- 
tion forming at a distance within the orbit-zone from the 
position of the planet's condensation, which was at the time 
beyond that of its prevailing attraction. Comets so formed 
may be termed planetary comets, and bear relation to certain 
planets, as Saturn, Jupiter, Mars. Short-period comets were 
also probably formed by local condensations outside the mean 
planetary plane at the same period as the planets were formed, 
when the planet's mass was not the superior attraction with 
respect to the position of condensation. They may also be 
formed by the detachment of parts of the tails of long-period 
comets disturbed at perihelion by disruption of the cometary 
matter by heat, which retarded the revolution-velocity of a 
part of the system. 

171. Symmetrical elements of Comet-formation. I have 
offered some general arguments for the construction and 
motions of local rotative systems formed in space and pro- 


jected towards the sun by its superior attraction, as suggested 
above, in a paper published in the 'English Mechanic' in 
1883 *. These ideas will be now reproduced, with some 
extenuations that appear to me necessary in reconsidering the 

We can scarcely enter into the discussion of the motions 
of comets without making a clear division of the subject as 
to whether they have been derived from purely nebular con- 
ditions directly, or have suffered from the attraction of other 
bodies near which they have passed so closely as to have their 
general motive symmetry destroyed. That this condition of 
disturbance occurs is quite evident in the present state of 
what was formerly Biela's comet, and in some comets which 
have from unknown causes become invisible, and may now 
only be represented by wandering meteoric systems, but 
whose orbits were clearly defined upon their first appearance. 
It follows, therefore, that for any symmetrical law of comet 
construction we must take into consideration such comets as 
retain the symmetrical form which we may suppose that they 
possessed when originally projected from space. These comets 
may be the only ones that are not under conditions of dis- 
solution, which may be at the present time the more common 
phase of comet life. If the comet depart through some 
disturbance from its law of original construction, its matter 
may present afterwards only what we may term a specialized 
confusion, too complicated to discuss by the most advanced 

172. The symmetrical comets will be presumed to possess 
elliptical orbits which have not been materially disturbed 
from the time of their original formation and projection 
towards the sun. Types of such comets may be found in 
Halley's, Donati's, the great comet of Sept. 1882, and many 
others of long period. They may be distinguished by possessing 

* ' English Mechanic,' 22nd June, 1883. 


a head and forward projection of the coma of symmetrical 
outline, after the manner of 7i, i, j, Plate III. 

173. It has been fully demonstrated that the head of a 
comet follows a truly elliptic or parabolic orbit ; so that we 
have no doubt that this part of the comet is subject to purely 
gravitational influences. The difficulty presented by these 
bodies is that the tail has not conformed to the conditions 
of orbit for the parts of a free system, or as it should do 
upon the swarm theory. This point will now be particularly 

174. Comets considered as Gravitative Matter. The cer- 
tainty that the orbits of comets conform to the laws of gravi- 
tation was clearly laid down by Newton as a principle. This 
would lead us to infer that they are composed of quite ordinary 
gravitative matter, which is again to a certain extent con- 
firmed by the spectroscope. The reason why it is thought that 
there must be a deviation from this law (by Olbers, Bessel, 
J. Herschel, and others who have followed this idea) is that 
during the perihelion passage of the comet, the tail, which 
must be considered a very material part of the cometary 
mass, diverges greatly from the normal elliptic or parabolic 

175. To meet this case, we have been asked to assume that 
the tail is unlike any form of matter with which we are 
acquainted, that it must be antigravitative, or that it becomes 
so from some cause at or near the perihelion passage of the 
comet. It may be suggested that the law of universal gravi- 
tation is one of the last we should abandon, seeing that it has 
done such perfect service wherever our knowledge of the 
conditions was exact. Further, there are other sufficient 
reasons by which we may conclude that there cannot be 
repulsion in any part of the comet, as the centre of gravity 
of the cometary mass follows constantly in the true orbit. 
For if a portion of the cometary mass, that is, the tail, changed 
its state of constant attraction so as to become unlike other, 


ordinary gravitative matter, and this portion became repulsive 
from the sun by heat, electricity, or otherwise, when the 
comet passed near perihelion, then the centre of gravity of 
the cometary mass must be altered by this repulsion, and 
would be displaced in relation to the sun's attraction ; so that 
the sun would no longer occupy the focal point in the orbit of 
the portion of the cometary mass that remained attracted to it. 
Therefore, a new form of orbit must necessarily be formed to 
suit the altered conditions of attraction, gravitatively central 
to the mass attracted only, or there must be in this case a 
deviation from Kepler's third law, a condition at least im- 
probable. As regards the change of form or properties of 
any known matter, so far as physical knowledge extends, we 
may take it that heat has no effect whatever upon its ponder- 
ability, so that difference of gravitation from this cause would 
not be possible through the difference of temperature induced 
in the comet by passing very near to the sun. Therefore, 
although the gaseous matter might be expanded, its centre of 
gravity would still follow as nearly as possible in the cometary 
orbit, not diverge. In other words, from the fact that no 
considerable change of orbit is evident after perihelion passage 
of a comet, we may conclude that the centre of gravity of 
the cometary mass traverses the true orbit of its projection in 
space, carrying with it the mass of the tail, and thereby that 
it conforms to planetary laws in the same manner as any 
other entirely gravitative system of matter. 

176. At the time Sir John Herschel suggested that the 
tails of comets might be of a kind of matter of which we 
have no knowledge, which is antigravitative in relation to 
the sun, the spectroscope had not been brought to bear upon 
cometary matter. We now know that carbon, hydrogen, 
and other elements form constituents ; so that the introduction 
of imponderable matter can scarcely be permitted, even hypo- 
thetically, in this case. Further, it has never been explained 
in this theory, how the cometary matter, after it is expelled 



from the sun, recovers its attraction or cohesion, so as to 
re-form the actual comet after perihelion as we know from 
actual observation that it does. If we adopt the theory of 
electrical repulsion as proposed originally by Gibers, and 
supported by Bessel, Norton, Zollner, Bredichin, and others, 
which is now most popular, this in no way relieves the 
difficulty. If the tail is repelled on approach of the comet 
to perihelion, with an internal separative force due to 
electricity of one sign assumed to exceed gravity, it must 
necessarily be left behind, and can never regain the orbit 
velocity of the head of the comet. Now this is precisely the 
opposite of what is requisite to represent the motion of an 
actual comet ; what is required is that the comet shall be 
elongated at perihelion, for which the action of gravitation 
alone is sufficient, and that the matter of the tail shall have 
its velocity of direct projection increased in such a manner 
that it shall describe larger arcs at the radii of its separate 
parts from the head of the comet, to which the head remains 
constantly as a centre during its perihelion passage. It has 
been suggested that the tail forms a small part of the 
cometary mass and may be re-formed from the matter of the 
head : this is in the highest degree improbable, as it is not 
necessary that a comet should even possess any head for 
that which represents the head is often merely the centre of 
inertia of the cometary system which follows in its orbit. 

177. Conditions under which a Comet may be considered 
as a Planetary Body. Taking the evidence of apparent con- 
ditions of comets generally, we find them immense volumes 
of what appears to be nebulous matter of somewhat sym- 
metrical outward form. Therefore, evidently forming systems 
of matter held together by internal forces which must in 
some way conform to the laws of gravitation, and so far 
resemble planets. We find comets otherwise of very small 
density, as is evident from their not disturbing the orbit of a 
planet whilst passing near it. Therefore, to account for such 


immense volume and small density, in a solitary or planetary- 
like system, we have in the first place to consider the 
possibilities of ordinary gravitative matter, which we now 
know it to be, being held together symmetrically by a system 
of forces, wherein the matter itself, although this is in a state 
of very great tenuity, remains practically an adhesive system. 

178. Now, following the analogy of things known to 
account for an enormous diffusion of matter from or about a 
central attraction of gravitation, being either engendered or 
maintained in an extensive nebula or planetary-like system, 
such as a comet may be considered to be ; we have only 
three known conditions which may so far react upon a 
gravitating or centralizing force in matter to insure this state 
of diffusion, tenuity, or decentralization : 

1. Heat-forces may separate the parts of a unit system of 
matter to any degree of tenuity. 2. Electricity of one sign 
may separate attenuated matter similarly to heat but with 
greater activity. 3. The tangential action of the revolution of 
the outward parts of a gravitative system about a centre or focus 
-may separate these parts proportionally to their velocities and 
distances from the centre according to the law of orbit to any 
degree of tenuity. 

If we consider the probabih'ty of the one or other of these 
forces being entirely or principally active in a cometary 
system, we find, with regard to the first, that to maintain a 
degree of heat sufficient to diffuse gravitative matter in a 
nebular form to such extreme tenuity as we witness in comets, 
we must assume great intensity of this heat, even if we 
assume the central attraction small. Further, for this heat 
to act as a separative force, we must assume a permanent 
gaseous state, as heated solids could not repel one another 
to produce the observed volume. Then, again, if we assume 
the heat present to be sufficient to account for the extreme 
diffusion necessary to produce the known tenuity, still we 
have the difficulty present that this heat will be subject to 



constant radiation in space, from all exterior parts of the 
system, and therefore the comet be subject to a constant 
loss of the decentralizing force, which alone in this case could 
support its tenuity. As we know that the larger comets pass 
to very distant regions, where little heat can be derived from 
the sun, we can scarcely imagine that heat sufficient to 
maintain the enormous diffusion of matter we observe can be 
sufficiently conserved in their entire systems under the 
excessive amount of radiation they must, experience in the 
clear cold regions of space. So that we must conceive that 
if the cometary state depended upon a force subject to such 
radiation, gravitation being constantly active within the 
system would, within a moderate period, reduce the comet to 
meteoric matter, which, owing to its reduced dimensions as 
solid matter, would after a period remain in this state, and 
become invisible to us, unless projected very near the earth. 

179. Further, if we assume the comet to be entirely 
gaseous matter, we can scarcely imagine a degree of internal 
heat in the system sufficient to render this, that is the hydro- 
carbon portion of it, visible, neither can gaseous matter 
reflect the solar rays. Therefore, upon the whole, we are led 
to consider the gaseous condition as highly improbable to 
account for the observed tenuity, and at the same time the 
illumination of the whole comet as a visible body. On the 
other hand, it does not appear improbable, with respect to 
certain comets, that sufficient heat is maintained in the 
nucleus (in some cases, perhaps, from passing very near the 
sun) to render this visible in itself, and sufficiently so also to 
illuminate the surrounding matter of the comet and tail to 
some extent within a certain distance from the sun. 

It is not necessary even to consider the nucleus a solid or 
liquid body. It may be a rotation-system composed of many 
parts reflecting light and yet transparent through the extent 
of the interspaces. 

180. The conditions under which electricity of one sign, 


+ or , could act within a unit system of matter such as a 
comet, are difficult to define ; electrical phenomena depend 
generally upon the tendency to establish equilibrium, but 
with our limited knowledge we cannot say a single form of 
electrical energy in a system is quite impossible. If possible, 
it may place diffusion of finely divided matter in equilibrium 
with the reaction of gravitation for decentralization of matter 
in a comet. At the same time, with electricity we stipulate 
a form of energy which is exhaustive, particularly if light is 
produced, in the same way that heat is exhaustive. So that 
we cannot imagine its conservation in the same manner that 
momentum is conserved in matter when moving in a friction- 
less medium. Therefore, all that we can say is that electricity 
probably takes a part in the phenomena of comets, but this 
would not account for their symmetrical forms. 

As regards illumination of the comet, the probability is 
that the electrical action present is a phenomenon exterior to 
the general cometary mass, exactly of the kind herein pro- 
posed for the illumination of the condensation of pneuma to 
nebula, by electrical discharge ( 42). 

181. We may now consider the third condition proposed : 
That the tangential action of the revolution of the outward parts 
of a gravitative system about a centre or focus may separate 
these parts proportionally to their velocities to any degree of 
tenuity. In this proposition we may observe that, if the 
cometary volume is maintained by the revolution of its 
outward parts moving in elliptic orbits, we have conditions 
that bear a strong analogy to the motions of planetary bodies. 
Thus, so far as we know, all planets and systems that we 
can observe are in revolution, both in their own masses, and 
in the attendant parts of their systems (satellites, rings). 
Therefore by analogy, taking the comet to be a part of the 
solar system, we must assume that the outward parts of the 
cometary system are in revolution. And as the satellite may 
revolve at any distance from its primary in an elliptic orbit, so 


also may any of the outward parts of a comet, however small, 
revolve. Further, if this motive system is assumed for the 
comet, we have the initial energy in the system conserved, as 
there is no loss as with heat radiation, unless we imagine 
resistance hy a surrounding medium, of which we have no 
evidence from other planetary or cometary phenomena. 

182. If we admit the conditions first suggested as regards 
heat or electricity to be in a certain degree active ; by this 
the nucleus of a comet may probably be heated or electrified 
liquid or solid matter in unit mass or compounded of many 
meteorites surrounded by a gaseous envelope, although this 
heat can scarcely be imagined to be sufficient to maintain 
the large outward cometary mass of the tenuity observed. 
Nevertheless, the nucleus would partly illuminate such 
exterior parts as I have suggested are in revolution about it, 
but in this we have clearly the necessity that such parts to 
be visible should be solid or liquid matter and not gaseous. 
Under this condition also the sun would illuminate the entire 

In comets that pass very near to the sun it is presumable 
that through the great heat the}'- receive near perihelion, any 
ordinary matter with which we are acquainted, and which 
might form part of the complete comet, would be reduced at 
the time to vapour or gas, in some cases possibly by explosion. 
This being the case, about perihelion, by radiation of heat 
received into space from the gaseous outward parts con- 
densations may occur in these parts about separate centres, 
again developing heat and electricity. The units of conden- 
sation may form sphericles under the same conditions as 
clouds are formed in the earth system. These would appear 
in mass as cloud, and in this state would traverse the orbit 
of the cometary system, forming parts of the coma and tail. 
If the head of the comet was in rotation, the condensed units 
would be in revolution at any extended position therefrom. 

Now, as regards the magnitude of each solid separate part 


or particle now suggested as an outward revolving part of 
the cometary system, this may be as small as we like to 
conceive it, assuming its mass sufficient to reflect a ray of 
white light. For the existence of such minute cometary 
matter, we may possibly find some analogy in the system of 
small meteors which revolve in elliptic orbits about the solar 
focus, and which are sometimes brought within the attractive 
distance of the earth, evidence of which is further given by the 
cosmic dust discovered in snow by Prof. Nordenskiold *. 

183. A form of particle which may be probable as a result 
of uniform condensations is that of a smooth bright metallic 
or vitreous sphere, which is covered with a permanent gaseous 
condensation, of hydrogen or a light hydrocarbon, which, 
under the sun's influence, may produce a gaseous envelope in 
the same manner as the earth is surrounded by its atmosphere. 
The refraction of the gas intensifies the sun's heat and light 
upon it. Refraction and reflection will carry over part of the 
light to other such globes at greater distance from the sun. 
Further, when the comet is sufficiently near the sun for its 
heat to render the gas, if hydrocarbon, incandescent in the 
purely isolated state, the comet may possibly become self- 

The change of state from the solid or liquid to the gaseous 
will also develop electrical conditions which may render its 
matter temporarily luminous when approaching the sun or 
receding from it by the effect of after-condensation. 

184. If the exterior cometary matter in separate particles, 
as proposed, is in rapid revolution in very elongated elliptical 
orbit round the nucleus of the comet, which I assume is 
necessary to maintain the extent of what we may term the 
cometary volume of the tenuity observed, then this revolving 
matter may be projected in elliptic orbits, either as separate 
particles (satellites), or more probably in accumulations or 

* Voyage of the 'Vega V 


series in one or more connected gravitative systems in the 
form of more or less perfect rings or bands, the orbits of 
which, about the head of the comet, may cut the solar- 
cometary plane at all angles. I will now endeavour to trace 
diagrammatically what would be the action of such a system 
whilst moving in an elliptic orbit round the sun. 

185. Comet-tails. Trains. Under the gravitative con- 
ditions proposed, the word tail will be entirely a misnomer, 
as the matter which forms the tail of the exterior parts in 
revolution about the nucleus is assumed to change position 
and become at another time part of the head. I will call the 
revolving parts of the head the pericoma, and the extreme 
of the tail the apocoma. For the entire comet except the 
nucleus, I will use the word train, which has been sometimes 
employed before. 

186. Following the conditions proposed, that the comet is 
made up of separate parts moving closely together in rapid 
revolution around the nucleus, we must then assume, by the 
laws of gravitation, that the centre of gravity of the mass 
held by mutual attractions will be constrained to follow an 
elliptic path for its orbit. But the separate parts of the 
system which together form the train may describe elliptic 
curves about their common centre of gravity, that is the nucleus, 
modified by their mutual attractions to each other , in combination 
with the attraction of the sun. This may be taken in detail. 

187. Elongation of the entire Cometary Mass near Peri- 
helion. Now, as regards the proper motion of the system of 
the comet constituted as suggested, we may consider that the 
velocities of its separate outward parts, as of all planetary 
systems including satellites, are such that the areas described 
by the radii vectores are proportional to the times that is, 
in relation to their own focus, and so far as this system is 
undisturbed by other attractions. Further, we know nothing 
in astronomy of resistance by a surrounding medium ; there- 
fore, however small we assume the exterior separate parts 


which are linked together by mutual attractions to form the 
train of the comet, these will separately maintain their 
velocities in relation to the nucleus consistently with the 
length of curves described by their radii vectores in equal 

188. These premises being granted, we may next consider 
the conditions present of the relative attractions upon the 
separate parts of a cometary mass in combination with that 
of its superior focus the sun. For this we may first, to save 
the complication of superimposed motions, consider the comet 
upon the Swarm theory as an immense system of matter in 
separate parts, either as a number of separate meteorites or 
cosmic or nebulous matter, connected closely together by 
mutual attractions but forming an elastic system. The 
matter exterior to the nucleus is assumed for the present not 
to be in revolution about the nucleus, as our theory demands 
that it should be. Then suppose that the whole matter of 
the comet is moving in its orbit. Take this diffused mass at 
the position of aphelion, at such a distance from the sun 
that the sun's attraction would not materially disturb the 
arrangement of its parts. It will then be clear that as the 
large cometary mass (now assumed not in revolution) 
approaches the sun, each of the separate parts of the mass 
will be accelerated directly as the length of curve described 
by its radius vector in unit of time about the solar focus. 
Therefore, the forward parts traversing space towards the 
sun will move at this approach much more quickly than the 
following parts, and the entire comet, possibly globular at 
aphelion, will be enormously elongated into ellipsoidal form in 
passing very near the sun. By the same laws also acting 
inversely in passing from the sun, its mass will be contracted. 
This principle is shown by fig. 16. 

Let S be the sun, and a b c three particles of matter in 
the train of the comet which is revolving in an orbit around 
the sun shown by the surrounding line. Let a be a forward 


particle, b a central particle, and c a following particle. 
Then will the velocity of a be greater than that of b at the 
period of approach towards the nucleus, and b greater than c 
proportionally to the lengths of curve described by their radii 

Fig. 16. 

vectores in unit of time ; so that as the comet approaches 
the sun its mass system or volume will be elongated in 
space in proportion to the excess of attraction to its forward 
parts by acceleration of gravity over its following parts. And 
under these conditions the greater the eccentricity of the 
orbit, that is the greater the acceleration at perihelion, the 
greater the length of the train of the comet. So that the 
outward form of the comet would in degree, so far as present 
conditions are considered, resemble its own orbit in form. 

189. Under the above conditions, we may observe that, 
upon the whole, in considering the cometary mass as not 
being in revolution, all the parts in any plane would follow 
each other in elliptic orbits about the sun ; and although the 
tail or following part would constantly increase in length in 
approaching the sun, it would not diverge from the sun in 
passing near it, or the separate parts move out of their separate 
solar orbits as gravitation units. Therefore there must be 
present, as generally admitted, other conditions to account 
for the divergence which is universally observed. In this 
proposition it is assumed that it is possible for the outer parts 
of the comet at greater distance from the sun to possess 
the same orbital velocity as the inner parts moving nearer 


to it, so that the figure of the comet may be conserved, as 
erroneously assumed by others in the discussion of the Swarm 
theory, in opposition to Kepler's third law. 

190. Orbits of the outward parts of a Comet. Focal 
Point. We may now follow the conditions just proposed, 
and assume that the comet resembles a planet surrounded by 
a connected revolving system of matter equivalent to a system 
of satellites, or to Saturn's rings, this revolving matter being 
meteorites, dust, or nebulous matter, and that the revolving 
mass (which I have denominated the train) will not only be 
sensitive to the attractions of its own parts and its nucleus or 
the centre of inertia of the system, but also to the attraction of 
the sun at the same time. Then, as the sun's attraction will 
not be linear with the major axis of tJie comet's mass in its 
assumed elongated form, the outward parts of the train, 
which are projected forward of the nucleus during their 
rotation, will be accelerated and be drawn towards the sun, 
and if the motive direction of these is inwards towards the 
sun, as it will be in the exterior part in revolution in the 
solar- com etary plane about the nucleus, these will be drawn 
towards the nucleus also; and in the smaller arc thereby 
induced by increased attraction will have their velocity 
accelerated (Kepler's second law) . In this manner, although 
the matter of the train will be elongated in space by the 
differences of velocity of its parts in passing over separate 
portions of the curve described by its radii vectores ; still the 
nucleus of the system will maintain a position at the forward 
focus of the orbit of the train, and parts of the train will be 
induced to move in an orbit round its nucleus, or the centre 
of gravity of the system, which will closely resemble that 
of the superior orbit of the entire comet in moving round 
the sun. 

This matter may be better conceived by reference to the 
diagram fig. 17. Let S be the sun, C the nucleus of the 
comet, a a particle moving in its orbit in the solar-cometary 


plane, shown by an elliptical outline. Draw a line through 
the centre of the sun, and through the nucleus of the comet 
to represent the linear direction of gravity on the orbit 

Fig. 17. 

shown by the outline. Then will the combined attractions 
of the sun and the nucleus exert greater attraction upon a 
particle in this position a as the comet approaches nearer 
the sun, than was exerted upon it at the same position of 
the orbit by the nucleus only, and such attraction will cause 
the particle to move faster and nearer to the nucleus, or 
inwards in a direction from a towards the point a'. In such 
a position its forward radii vectores in relation to the comet's 
centre of inertia will be closed, and its velocity increased 
proportionately to the lengths of curve now described by its 
smaller radii vectores. It will thus also maintain the nucleus 
at its focal point within the comet, although the cometary 
mass be elongated by the increased velocity of its forward 
parts at pericoma as shown above. 

191. Direction of the Cometary Train in Relation to the 
Sun. By the above conditions it will be seen that the 
separate particles of the train of the comet will be actuated 
by forces which will be the resultants of the combined 
attractions of the sun and those of their own proper nucleus, 
the pericoma of the train, if this principle is admitted, moving 


about its head will come constantly in the direct line of 
attraction between the sun and the head, and will constantly 
change the direction of the attractions upon the parts of the 
train. The particles of the train will also be subject to the 
attractions of their own masses upon one another ; and if 
the train is a combination of dust or meteorite ring-systems 
as suggested, these attractions will still further modify the 
conditions given. The principle now offered may be best 
described diagrammatically by fig. 18, for which I will again 
take the position of revolving matter on the solar-come tary 
plane only. 

192. Thus : Let S be the sun, C, C', C", fig. 18, three 
positions of the nucleus of a comet moving in the orbits 
Y Z, Y' 71, and Y" Z". Let d e be two particles of matter 
which form part of a continuous attractive series constituting 
the train. When the nucleus is at C, and moving in the 

Fig. 18. 

orbit Y Z, then will the continuous series of particles arriving 
at d be not only actuated by the attraction of their nucleus, 
but will be accelerated by the attraction of the sun directly 
linear to their motive path in the direction d, as before shown 
a to a! in fig. 17, and all following parts as they arrive will 


be mutually attracted in series in the same direction. There- 
fore the particle d, fig. 18, will be both accelerated, and be 
drawn at the same time towards its nucleus C, and by the 
curve of its smaller radius vector it will pass nearer the nucleus 
with greatly accelerated speed. 

193. Now, as regards the attraction of its nucleus and the 
velocity engendered in particles moving about it by the 
attraction, the velocity would be maintained, varying only 
in inverse proportion to the squares of length of its radii 
vectores onwards to e ; but from d to e the sun's attraction 
would still draw the particles towards the cometary nucleus, 
so that throughout perihelion passage the curvature would 
be in a certain degree maintained as at perihelion, directing 
the particles of the train thereby towards the point x. The 
apparent effects of this will be that the whole mass of the 
train considered as an elliptic mass in revolution will be, as 
it were, pulled forward in the direction of its rotation, and 
whirled round in space so that the position of the suns centre 
may be maintained constantly linear with the least radius 
vector of the parts of the train that pass nearest to it, and 
the whole train of the comet, although describing an ellipso- 
epicycloidal curve, will be as far as possible constantly 
symmetrical about the nucleus ; the orbit of the train thereby 
changing to make this possible from the positions Z Y to 
Z / Y ; , Y" 71' when the nucleus of the comet arrives at positions 
C, C', and C". 

Under the above-stated conditions the revolving matter 
of a comet passing its perihelion will possess much greater 
absolute velocity than the centre of gravity of the comet. 
Therefore a comet may pass very near the sun's surface with 
momentum too great from this self-rotatory velocity for its 
matter to be materially disturbed by slight resistance of 
attenuated matter, if such exists about the sun. 

194. Widening and Curvature of the Train by crossing 
Orbits. One other point to be considered under the conditions 


given above is that the newly placed pericoma at every 
change of position in relation to the cometary nucleus, as it 
becomes directed towards the sun, will influence the parts of 
the train only in proportion as they are accelerated. There- 
fore, the parts of the nucleus which have just passed the 
pericoma will have the accelerative force due to the sun fully 
impressed upon them, whereas the parts arriving there will 
not be so fully impressed, but will retain a part of the force 
due to their initial velocity in relation to the previous position 
to the nucleus at the last time they passed between it and 
the sun. Therefore, the following parts will not maintain 
quite a symmetrical position with respect to their future 
pericoma, but will lag behind this position, so that the general 
path of projection of the train will be constantly of greater 
curvature upon the exterior of the orbit than in the parts 
nearer to the sun. 

195. If the matter in revolution about the cometary nucleus 
revolve in different periods according to Kepler's third law, 
as the planetary matter about the sun does, then the orbits 

Fig. 19. 

of the parts of least period of revolution, which will be nearer 
to the nucleus, will appear to lag less than the outward 
parts of longer revolution period about the nucleus, so that 


by this means a crossing of orbit will occur, producing either 
a widening or opening of the tail. Under these conditions 
also possible collisions may occur, producing complicated 
effects impossible to follow here. Fig. 19 represents the 
normal conditions in the solar-cometary plane where the 
crossing orbits are assumed to be conserved. 

196. Let S be the sun, C thecometary nucleus, ^represent 
the orbit of a particle of early revolution, e the orbit of a 
particle of later period. After perihelion of the entire comet 
in relation to the sun, this system of forces, by the changed 
dispositions of attractions, will act inversely. 

197. Formation of a neiv Head or PericepJialion about 
Perihelion Passage of the Comet. Under the above conditions, 
independently of the complication of motion of cometary 
matter moving in orbits not in the solar-cometary plane, the 
conditions of which have not been separately considered, 
incidental phenomena must occur which will, more or less, 
disturb the general conditions thus : As the sun causes 
acceleration to the parts of the train in approximating their 
pericoma in relation to the head, and retardation to these 
parts in leaving it, the matter of the train will be more con- 
densed round the head in the point of pericoma of the system, 
and thus form incipiently a new head, which will react, by its 
gravity, as a secondary or false focus, and in its turn, by 
maintaining the centre of gravity in the cometary orbit, will 
tend to disturb the position of original matter that formed 
the original focus. At the same time, if the comet passes 
very near the sun, the head, by the great heat it receives, 
will be expanded or even exploded ; so that it may become a 
less dense mass ; or even new heads may be again formed 
forward of this with a general relative disturbance of the 
internal gravitation in the parts of the cometary system, the 
conditions of which are too difficult to follow. 

198. It is at this point where, as pericomic matter increases 
in density at the head of the comet, and heat forces, and 


possible development of electricity, cause great increase of 
internal elasticity, the orbits of train-matter may be separated 
by explosion, so that a comet may be divided into two or 
have a large part of its train detached from its own focus, 
one part to be retarded in its orbit as much as the remaining 
part of the comet will be accelerated mass for mass. Or the 
complexity of orbital conditions induced may separate a part 
of the train about perihelion where solar action is most 

199. Some General Conditions. As to certain appearances 
of comets under the conditions proposed, which will influence 
the spectra obtained therefrom if the nucleus of a comet is 
a heated or electrically excited mass, which it may become 
by the collisions of parts of the train in revolution near 
pericoma, or by heat derived from the sun. Under the same 
conditions the train in passing pericoma may be also heated 
or electrically excited sufficiently to become luminous or 
phosphorescent. This luminosity will gradually die off in 
the distance, so that possibly the parts of the train of a comet 
in aphelion are seldom visible. From which we may con- 
clude under these conditions that visible comets are generally 
only the parts of the cometary system that are nearest the 
nucleus. In the comet of 1882 (Wells) I was anxious to 
observe whether any appearance could be detected of orbital 
projection beyond the extent of visible tail, which could be 
possibly observed by some obscuration of part of the faint 
light from celestial space circumscribing a curved elliptical 
space. This partial obscuration I fancied I observed on the 
23rd October at 5 A.M, It was also observed as a darker 
part behind the visible tail by Mr. C. J. B. Willians at Cannes, 
France*, as also by Mr. Henry Cecil at Bournemouth f. 

200. Upon principles here discussed the cometic matter 
in revolution about the nucleus causes each particle to be 

* Nature, Oct. 26, 1882, p. 662. t Nature, Nov. 16, 1882, p. 52. 



illuminated by the heat or electrical excitation engendered 
through constant collisions, also by the heat of the nucleus, 
and by that of the sun on the side turned towards it. There- 
fore the light we receive by reflection will be proportional to 
the amount of illuminated surface of the revolving particles 
visible to us, and the direct light that of incandescence. In 
this manner the nucleus itself may become invisible or dimmed 
by eclipse of revolving incandescent matter about it. On 
the other hand, the light received directly from the nucleus 
will be proportional to the open spaces only between the 
outer revolving parts crossing the field of light. From these 
causes it is possible that some of the most remarkable appear- 
ances of projection of cometary matter suddenly from the 
nucleus through immense distances may be merely the 
lighting of the distant parts of the train of cometary matter 
through interspaces of the central system upon dispersed 
matter which, although present, was previously invisible 
to us. 

201. The matter of some comets, of which Encke's is 
perhaps one, may be a carbon-hydrogen compound, which in 
the distance of space, as at aphelion, may be condensed to 
particles of solid matter, but in nearing perihelion may be 
again converted into gas by the heat of the sun with develop- 
ment of incandescence through electric excitation sufficient 
to render it visible. This matter may again condense in 
passing towards aphelion, the orbit position of the separate 
units of the system remaining the same with the electrical 
effects of change of state the apparent outward visible 
volume of the comet varying according to the conditions 

[ 147 ] 



202. The Earth may be considered as a model planet 
whereon we are able to observe the evidences of exterior 
conditions which are due to phenomena that have acted upon 
it to produce its present form and constitution. In this study 
we may possibly approach the conditions which also ruled 
in the formation of all the dense planets interior to Jupiter, 
of which we can possibly obtain no further evidence of struc- 
ture than that apparent upon the surface of Mars and the 

203. To assure our premisses for earth-structure it will be 
convenient briefly to recapitulate some general propositions 
that have been already discussed, which we may possibly 
accept as data for certain factors of early formation. The 
most important of these are : 1. That the solar-planetary 
system during condensation possessed more or less nebular 
matter projected about its equatorial zone, as represented 
diagrammatically by the discoid form in fig. 9, p. 67. 2. That 
the tenuity of the planet-forming system interior to Mars 
upon the condensation of the sun's volume, was too great to 
support the nebulous concrete state at a period when the 
exterior solar nebula was falling below its critical temperature 
( 112). So that within the orbit of Mars local separate con- 
densations were formed at first of the more refractory matters 



widely exterior to the earth's orbit. 3. That at the early 
period when the intra-Mars local condensations were forming 
the earth's nebula was still attached to the sun ( 146) . 4. That 
the exterior local condensations of matter at an early period 
were drifting sunward into the earth's nebula and towards its 
orbit position in moving under the resistance of the surround- 
ing less refractory nebulous matter. So that the earth during 
its formation at no time extended in an entirely uncondensed 
nebular form so far as the orbit of Mars. 5. That the 
earth's formation being partly but not entirely due to the 
condensation of gaseous matter as assumed herein of the 
planets Jupiter and Saturn a denser system was produced. 

204. Under the above-stated conditions we have the 
suggestion of two large factors of earth - formation : An 
interior nebular system about the earth's orbit of purely 
gaseous elements, which were sinking at the period of forma- 
tion to a critical point of temperature and impressing their 
virtual momentum upon the earth in the direction of its 
rotation, upon principles already discussed 128, and an 
exterior condensation system of meteorites or planetoids which 
were projected into the earth's nebula. These, as bodies 
moving in nearly free orbits, that is, under slight resistance 
of the residual nebulous matter, drifted into the earth's nebula 
in spiral orbits that upon contact with the then forming earth 
may have produced effects which continue to be evident 
upon its surface. 

It is necessary in this matter to insist upon the con- 
tinuance of nebular conditions during the greater part of the 
period of the formation of the mass of the earth, as such con- 
ditions alone could have produced its present direction of 
rotation, as it was shown originally by Descartes and Laplace 
( 7), and with equal clearness by M. Faye ( 141), that if 
it were formed from an entirely discrete system of matter 
moving in free orbits its direction of rotation would be the 
reverse of what it is. 


If we oinit from consideration the projection of discrete 
matter into the earth's nebula, which would carry with it 
certain elements of relatively retrograde momentum accord- 
ing to the theorem of M. Faye, the necessary extent of nebula, 
if this alone was active, may be inferred by taking the 
formula 27T7' 2 D = 2 TT(I\ r), from which we find T)r 2 + r=r ly 
D being the days in a year, r 2 the radius of the earth, r the 
radius of the earth's orbit, r t the outer radius of the nebular 
ring, which by calculation to produce the present rotation is 
found to be of about 1,000,000 miles greater radius than the 
earth's orbit. 

205. If while the earth was moving at nearly orbital 
velocity there were projected upon it certain factors of discrete 
condensation, giving a momentum the reverse of that due to 
direct condensation of gaseous matter, the radius of the 
original gaseous matter would have to be increased in pro- 
portion to the momentum of the discrete matter in order to 
account for the present rotation period. 

If we accept the conditions proposed above, they entail 
certain relative consequences of which we should expect to 
find evidences in earth-structure. 

206. Firstly : If the earth were condensed from a gaseous 
system, this condensation could not have been effected without 
producing an intense heat in the condensed matter, as it may 
be supposed to have resembled our sun in constitution at an 
early period. Further, such heat as would be produced by 
gaseous condensation must have also rendered any discrete 
matter which may have been projected therein liquid. 

In the formation of oxides upon the earth, if these were 
produced from elementary matter, the only probable condi- 
tion, they must also have produced great heat during their 
oxidation. At the same time, as such oxides are lighter than 
their metallic bases and are non-conductors of heat, they 
must have rested upon and covered the surface and conserved 
the central heat. Therefore we demand, ' in the first place^ 


upon these conditions, evidences of a highly-heated liquid 

207. Secondly : If the planetoids, large or small, produced 
by the condensation of matter exterior to the earth's nebula were 
incorporated therewith when the earth became a liquid planet, 
then such matter within or upon the earth's surface might pos- 
sibly remain geologically evident. Particularly such planetoids 
as may have drifted to the earth's surface when all the denser 
matters of its nebula were condensed. Under these conditions 
it remains probable that some land-areas of the earth may 
show evidences of this discrete form of condensation. These 
conditions are so far actual that we have still planetoids, that 
is, meteoric matter, falling to the earth. 

208. The nebular conditions which upon the theory herein 
proposed would have been constantly active at an early 
period of earth-formation require consideration in detail. 
They will therefore be deferred to another chapter with the 
exception of the argument upon which they must be supported 
of the entire internal fluidity of the earth. This also is 
necessary for the consideration of the effects of later discrete 
projections upon the earth to be discussed in the following 
chapter, as I suggest that these discrete projections were 
coincident with nebular condensation and may remain, at 
present, evident in land-formation. 

209. The Internal Fluidity of the Earth. This could not be 
questioned for a moment if the evidences of observation were 
alone considered, but in the contemporary science of any 
period we have generally in popular learning a tendency to 
depart from concrete observations directed to consider some 
hypothesis or isolated fact, which is elevated to predominance 
above all the natural conclusions of otherwise universal 
observations. I remember, when a cool sun was the prevail- 
ing theory, arguing with a professor that the intense heat of 
that body was manifest in many ways. His reply was that 
that was nothing to the purpose if the solar spots were found 


to be hollow places or cavities in the sun's surface ; as they 
were dark they must be cool ; and science must take account 
of every phenomenon. If experience teaches us anything, it 
is that theories change with every phase of science ; so that 
our safety, if we are seeking truth, lies in taking the mean 
evidence of all relative phenomena, not of any single pheno- 
menon which we may or may not perfectly comprehend. 

The most important evidence of the former nebular 
condition of the earth besides its direction of rotation is to 
be found in its interior condition ; for it is certain, as just 
stated, that the matter which forms the earth, if it pre- 
viously existed in a gaseous or vapourous state, must have 
been condensed from this state to a liquid with great 
development of heat. The density of the system also clearly 
indicates that the internal matter, if at a high temperature, 
is most probably metallic. Now as it is the property of 
metals to alloy and also to conduct heat, it is in the highest 
degree improbable, as sometimes supposed, that any part of a 
uniform system of condensation, as that of the earth, could 
be for ages intensely heated in some interior parts, and cool 
or solid in other parts. It therefore becomes a rational con- 
dition of the nebular hypothesis that the interior of the earth 
was, or is, in a highly heated uniformly liquid condition. 

210. If the earth at an early period condensed refractory 
metallic matter to form an intensely heated globe, while this 
was surrounded by less refractory matter, and particularly by 
oxygen, near the surface, then the globe would be formed 
mainly of the refractory matter, and be covered with the 
oxidized matter upon its further condensation. The oxidized 
matter must necessarily have rested upon and outwardly 
covered the metallic matter, so that possibly at a very early 
stage of condensation there was a metallic uniformly heated 
globe covered with a coating of non-conducting oxidized 
matter as we find it at present, although this coating at an 
early period would be very much thinner than it is now. 


As the earth cooled the coating must have increased by 
condensation and protected the earth more and more from 
radiation of internal heat until it was possible for water to 
rest upon the earth ; so that the heat of the interior must have 
been protected by non-conducting material, highly heated 
from chemical combination, in such a manner that it could 
only very slowly radiate its heat into space. 

211. We now arrive at an important point. Is the inte- 
rior of the earth fluid ? From mathematical physics alone 
in the accepted theory of the tides, Lord Kelvin, our greatest 
authority, says that it is solid, or the tides would not present 
the great variation of height of surface which we witness. This 
matter is repeated with authority in the recent compilations 
of Lord Kelvin's works *. There is one point of this theory 
at least which astronomers ought to be able to solve. Lord 
Kelvin shows that the necessary result of tidal friction is 
that the earth's rotation must be decreased by a total value 
of twenty-two seconds in a century ; and the establishment of 
this as a fact might tend to prove the certainty of the effects 
of tidal friction upon the earth, and at the same time possibly, 
if we had no other direct observations to consider, of the earth's 
solidity. If we depart from physical theory and follow the 
evidences of astronomy by observation, wherein the data are 
results of experience not liable to receive much correction 
from change of theory, we have undoubted authority from 
observation that the rotation period of the earth, considering 
the moon and Mercury particularly as time-keepers, has not 
decreased by a single second in a thousand years. 

212. If we follow the evidences of geology founded upon 
observation, the solidity of the interior of the earth appears 
to be quite impossible unless we assume an unknown form of 
matter which does not become fluid at a high temperature. 
We need only cite a few instances to show the improbability 

* Prof. 0. J. Lodge. Keview in < Nature,' July 26, 1894. 


of the rigid solid earth proposed by Lord Kelvin. In 
every considerable volcanic eruption, even within modern 
times, the matter that issues from the earth is at a white heat; 
and although this may be reduced to dust, as in the Krakatoa 
eruption, the microscopic examination of the dust shows 
clearly by structure that it was formerly, or just before its 
issue, at this heat and in a liquid state *. We find also that 
this matter may again be reduced to liquid at a heat much 
Jess than that at which it was thrown from the volcano. The 
mass of matter projected from Krakatoa is estimated to be 
equal to about 22 solid miles. The eruptions of other moun- 
tains in Java were of much greater mass. The liquid lava 
that issued from Skaptar Jokull in Iceland, in 1783, is 
estimated at 21 cubic miles. The lava of the Snake river, 
Idaho, North America, that has issued in tertiary times is not 
of much less volume than 500,000 cubic miles. We have 
also upon the surface many thousand cubic miles of lava in 
Abyssinia and in other parts that has certainly issued at a 
white heat. Further, the axes of our mountain chains are 
built up of plutonic matter the structure of which shows the 
evidence of former intense heat with extrusion of liquid 
mineral matter. This is evident although the surface-matter 
is often dislocated and rearranged so as even to include 
superficial sedimentary rocks, the deep-seated felsitic rocks 
seldom intruding beyond the base of a volcanic chimney. 
The volcanic and sub-volcanic matter, therefore, upon the 
earth's surface, which has evidently been in a liquid state, 
amounts to many millions of cubic miles. When we consider 
the wide distribution of this matter, its nearly uniform mineral 
structure, and its present almost constant issue, as in Kilauea 
and other volcanoes, we cannot but conclude that it formed, 
and now forms, a part of a general system whose intense heat 

* See author's paper, ' ' Krakatoa Dust/' R. Met. Soc. Quart. Journ. 
vol. x. p. 187 (1884). 


is derived from the uniformly heated state of the central 
matter of the globe. 

213. The evidence of the thermometer, which shows that 
there is a general increase of temperature with depth of ahout 
1 Fahr. for every 60 feet, where freedom from the influence 
of percolated water permits this measurement to be made, 
gives an increase of temperature of 88 per mile ; so that at 
about 200 miles we should have, at this rate, a temperature 
of 17,600, which no known body could bear while retaining 
a rigid state. It is not, however, necessary to consider a 
constant increase of heat if the central volume is metallic, as 
herein proposed, for a good heat conductor would distribute 
heat equally in the interior. The following suggestions may 
be offered as regards the internal liquidity of the earth. 

Fig. 20. 

214. Is the assumption of the solidity of the earth neces- 
sary to account for tidal action ? Having devoted some years 
to the consideration of motion in fluids, I find that the mobility 
of liquids depends greatly upon the freedom of surface, which 
adapts it to offer certain forms of accommodation for motion 


for which time is required. So that a dense liquid held in 
equilibrium by surrounding pressures, such as we may 
imagine the interior of the earth to be, offers considerable 
resistance to deformation through molecular friction in 
moving from a state of relative rest until certain forms of 
accommodation can be brought about, which are nearly im- 
possible in a close system. In firing an Enfield rifle bullet 
directly upon the surface of water, the point of the bullet does 
not pierce the water, but the water pierces the bullet, and 
reduces it to a thin conical shell *. Fig. 20 shows a full- 
size section of the bullet. Fig. 21 was taken directly from 
this bullet fired from an Enfield rifle normal to the surface of 
water through a thin piece of bladder in the end of a deep tank. 
The point of the bullet was painted red, and left to dry. 
The colour upon the point was found spread out to the ex- 
treme circumference, where it formed a curled-up edge, as 
shown in the figure, a a a. The centre of the figure shows 
the pressed out diaphragm of the bullet, which was reflected 
from the surface of the water, so that it forms no part of the 

The velocity of the earth's surface at its equator is much 
greater than that of the bullet; therefore we have to consider 
accommodation in the waters of the ocean and of the liquid 
interior for displacement in relation to time, in order that it 
may possess a certain form of possible motion of accommoda- 
tion. This presents a difficult problem, the factors of which 
are, for the greater part, quite unknown. If, on the other 
hand, the deformation under the moon's attraction is considered 
to depend upon the elasticity of the system, then a solid is quite 
as elastic as a dense liquid under great compression, so that 
neither solidity nor liquidity could be inferred from this 
cause. Gold or steel in a solid state is much more yielding 
under pressure in confined space than water, as is proved in 

* 'Fluids,' p. 187. 


the coining of metals, in contradistinction to the compression 
of water in the hydraulic press. 

215. Judging from the constitution of meteoric matter 
that has fallen upon the earth, assuming cosmic matter in a 
certain degree general, the centre of the globe would be 
largely composed of nickeliferous iron. As regards this 
matter, we have not been able to produce any temperature in 
our furnaces high enough to melt it, so that assuming this 
formed the larger part of the interior of the earth upon which 
the surface rocks rest, the white heat of volcanic matter as it 
issues from the surface of the earth, taken as an index of the 
interior temperature, would only be sufficient to raise pure 
iron or nickeliferous iron to a stiff' plastic state, as iron in 
forging offers very great resistance to change of form under 
great surrounding pressure. 

216. It is certain that the earth's rotation upon its axis is 
combined with that of the rotation about the centre of inertia 
of the earth and moon ; therefore there must be a swing in 
the free surface of the ocean, due to its plus and minus daily 
and monthly rotation-velocity in relation to its position with 
respect to the moon, which must produce tidal action in the 
free surface water. But how far this plus momentum in one 
part is compensated by the minus momentum in another may 
be a difficult problem to solve. One must, however, feel in 
this, as in other instances with which history furnishes us, that 
in the intent observations and calculations of certain actions 
we are liable to lose sight of the reactions that are not super- 
ficially evident or may be unknown. So that it becomes a 
question whether the motions within and about the earth 
moving in its orbit in frictionless space seriously affect its 
general momentum of rotation taken in a wide astronomical 
sense, so that we may infer its internal condition therefrom. 
Certainly, taking the matter in its entirety, we cannot suppose 
it to deviate greatly from Newton's Third Law, Cor. iv.: " The 
common centre of gravity of two or more bodies does not 


alter its state of motion or rest by the actions of the bodies 
among themselves." 

217. Effects of Tidal Friction. Taking account of the 
suggestions offered in 153 for comparing the revolution of 
the moon with the rotation of the earth upon nebular con- 
ditions, the effect of tidal friction upon the present rotation 
of the earth must have been very small, the proof of 
which is quite evident in early geological stratification. That 
there have been certain effects is suggested to be made evident 
by the present acceleration of the moon of 10" or 11" in a 
century. This effect was suggested by Kant and afterwards 
worked out mathematically with varying results by many 
eminent astronomers. Laplace accounted for the acceleration 
by decrease of ellipticity of orbit. These calculations being 
revised by our late eminent astronomer and mathematician 
Adams with greater precision, reduced the Laplace factor by 
6", leaving a residual 4" or 5" to be accounted for by tidal 
friction or some other cause *. Whether this residue may be a 
recurring differential from astronomical causes remains to be 
proved. If we take it to be a constant of acceleration due to 
tidal friction, then subtracting the Laplace-Adams factor, 
and assuming the earth's angular rotation-velocity to have 
been retarded in the same ratio as the moon's angular revo- 
lution-velocity was accelerated, and considering 5" per century 
as the mean acceleration, we have a period of about 12 millions 
of years for the time that the angular velocities of the earth 
and moon were equal ; that is, the condition assumed herein 
of the purely nebular state before the separation of the earth 
and moon, which must have been followed by a long period 
of time before there could have been any geological conditions 
that could remain evident in the structure of surface rocks. 
This short period can, therefore, scarcely be discussed if we 
take any recognition whatever of geological evidences. 

* Phil. Trans, vol. cxliii. p. 397. 


218. Change of Figure of the Earth due to Rotation-velocity. 
If there are sufficient scientific data to assume that our pre- 
sent day is longer than formerly, of which there appears to 
be a probability only within very narrow limits, then this, so 
far as concerns the present investigation, points to the con- 
clusion that the mean symmetrical figure of the early earth 
as a spheroid of revolution must have been changed by this 
condition and have been at the time of greater rotation- 
velocity more oblate. Upon the nebular condition just dis- 
cussed, the difference could not have been great. The effects 
of contraction of the equator under decrease of velocity, if 
we may take it at the extreme value accepted by some 
scientists, was ably discussed in a paper read by Mr. Wm. 
B. Taylor * before the Philosophical Society of Washington, 
May 3, 1885. 

219. Mr. Taylor takes the very extreme condition proposed 
by Prof. G. H. Darwin of a rotation period of six hours, which 
must have produced an elevation of land at the equator one- 
tenth greater than at present, assuming the earth to have been 
a true spheroid of revolution. He finds that this would make 
the equatorial radius 4359 miles and the polar radius 3291 
miles only. The pole would be therefore 658 miles nearer the 
centre and the equatorial protuberance 396 miles higher than at 
present. Mr. Taylor suggests that this would account for all 
the crumpling and inclination of strata and the elevation of 
mountain-chains. It is very doubtful whether it would do 
so ; the mean inclination of strata at all depths is probably 
not less than 20, and this could not be produced even with 
one tenth the equatorial contraction, supposing the inclina- 
tion in the past constantly increased instead of oscillating 
locally up and down during all geological time, which must 
have been the case from local volcanic disturbance that is quite 
evident in extreme cases by some strata being quite over- 

* American Journal of Science, 3rd series, vol. xxx. p. 249. 


thrown and inverted. Nevertheless it may be much more a 
vera causa than the contraction-theory of the late Robert 
Mallet. It may have been a cause of elevation of land in 
the tropics as an early condition, but it must have been 
materially modified by the condensation of nebular matter 
at the poles which I shall propose further on. 

220. The effects of the rate of rotation of the earth upon 
its oblateness appear to have been first suggested by the 
Rev. 0. Fisher, and carefully considered; but he regards them 
as very slight. In his interesting work on the ' Physics 
of the Earth's Crust ' he says : " The friction of the tides, 
whether oceanic or bodily, must necessarily have diminished 
the rotational velocity and lessened the oblateness. The parts 
of the crust about the poles will have been subjected to 
stretching and those of the equator to compression. There is, 
however, no apparent reason immediately to connect the 
inequalities with this cause, for the continents do not occupy 
an equatorial belt, as they would do under this hypothesis, 
nor have the polar regions been free from the compression 
which all continental areas have experienced " *. This fact 
appears to me to be a sound objection to the whole hypothesis 
of a former rapid rotation of the earth, seeing that the known 
geological evidences of early stratification are directly opposed 
to it. 

* ' Physics of the Earth's Crust,' Chap. xiv. p. 183. 

[ 160 ] 




221. Formation of Land-areas by inclusion of Planetoids 
into the Eartlis Surface. In this chapter it will be convenient, 
in order to simplify the subject, to take the conditions which 
were proposed secondly ( 207) of the factor of earth-formation 
from planetoid matter, deferring the more important con- 
sideration of nebular formation until the next chapter. The 
fall of meteoric or planetoid matter which may have been 
formed originally by condensation in a part of the space- 
interval between the earth's original nebula-zone and Mars 
will be now considered. 

222. Assuming the discoidal system of our sun's nebula 
shown in fig. 10, p. 85, we should have between the earth and 
Mars a thin attenuated nebular plane, in which local conden- 
sations would occur similar to that before suggested for the 
system of asteroids ( 112). Under certain local conditions 
such condensations may have assumed any possible dimen- 
sions of mass according to the amount and disposition of the 
surrounding nebular matter and to its initial motion being 
directed so as to form a centralized system or otherwise. 
Upon these conditions any local condensation would form 
separately what we may term a small planetoid, or merely an 
isolated unit of impalpable dust. 


The units of discrete solid matter formed within the space 
above defined, capable of composing any part of the future earth, 
must have possessed at the time of their formation a revolution- 
velocity less than that required for a circular orbital motion, 
according to the law of orbit. Otherwise they would have 
maintained their orbits and still circulate round the sun, a 
possible condition at present for many of these condensations. 
If the condensed matter was endowed with a velocity less 
than the above, as before stated, the sun's attraction would 
draw the early condensations into elliptical orbits ; and 
consequently such as had a perihelion distance from the sun 
within the distance of the outer periphery of the earth's nebula- 
zone at the time must have fallen into this zone and have 
combined with it, through the resistance which such projection 
would encounter in passing into or through the nebula-zone. 
In this case it is easily seen that solid matter projected into 
the earth's nebula-zone must have combined motively with 
the momentum of the matter of the zone, and, if it retained 
its distinct solid condition, fall in spiral paths sunward, unless 
it was drawn by a stronger local attraction within the zone- 
ring or the nebulous globe of the incipient earth. 

223. The period when exterior condensation would 
intrude into the earth's nebula-zone would depend upon 
many conditions. If its perihelion passage was entirely 
resisted by the zone it would enter a part of its system ; but 
if the resistance was insufficient, on account of the small 
density of nebular matter, it might continue its projection 
with an orbit of changed ellipticity in any degree. If it 
arrived at perihelion within the zone-ring after the ring had 
condensed into globular planetary nebulae or into a single 
nebular globe, its perihelion passage might strike or miss 
this nebula, or later miss the new-formed earth upon its 
further condensation in such a manner that there may be 
remains of intra-Mars condensations still falling as meteorites 
across the earth-orbit owing to want of former coincidence of 



perihelion position with the earth's place at the time of 
internal conjunction. 

224. Under the above-stated conditions, during the discrete 
condensation of intra-Mars nebular matter and its projection 
earthward, so far as this could overcome the resistance of the 
nebulous matter to continue its projection in a solid form, it 
would form earth-surface matter in combination with the 
nebular earth-zone which would still be forming. By this 
entire effect, the surface of the globe would receive a mixed 
condensation of matter upon its surface composed of the 
discrete matter, which would contain every form of solid 
elementary matter, and of the residual nebulous matter that 
was condensing at the period. 

The above-defined mixed conditions of deposition correspond 
fairly well with actual observed conditions, as it is clear that 
we have no regular density system such as would be pro- 
duced under the condensation of a purely nebular system. 
We have gold, copper, iron, and other dense metals upon the 
surface, which may at an early period have formed the central 
systems of discrete condensations from the universal pneuma 
before their projection to the earth. 

225. In the small amount of exterior matter that has fallen 
upon the earth in recent times we have been able to recognize 
the presence of 23 elements, for the greater part those that 
generally prevail upon the earth, but in a few cases some of 
its rarer elements. They show upon the whole that they are 
condensations from an original universal nebula, in which the 
heavier as well as the lighter elements appear, iron largely 
predominating. The mean density of the entire mass of such 
meteors as have fallen to the earth in modern times is possibly 
not far different from the mean density of the earth, about 
5*6 times that of water, which should be the case under 
purely open nebular conditions of condensation from a 
universal pneuma system. 

It cannot, however, be suggested that the presence of 


heavy metals upon the earth's surface is entirely due to its 
reception of meteoric matter. There are other sufficient 
causes made quite evident in special cases. Metallic veins 
appear in many cases to be caused by the evaporation of 
metals from the heated interior, through fissures which were 
produced by the upheaval of primitive and later rocks. This 
evaporized matter is either in the form of pure or alloyed 
metals or of haloid compounds often modified by the presence 
of heated water and possibly originally combined in gaseous 
emanations. There is another cause to be proposed further 
on, namely the effects of the loading of ice at the poles upon 
the central system of the globe, causing projection of interior 
matter; so that the formation of land by the inclusion of 
meteorites may not be thought to be even a necessary 
condition of its formation, but only a very probable factor. 

226. To recapitulate the condition proposed that possibly 
will agree with observation of geological structure : We 
may assume that after the formation of Mars at the limits 
of the sun's nebular periphery, planetoids were again formed 
somewhat similar to our asteroid system existing exterior to 
Mars, from condensations due to the reduction of the tem- 
perature to the critical temperature of the sun's nebula within 
a thin peripheral plane, M, E, fig. 10, p. 85. We may assume 
that these planetoids were formed at first of the more 
refractory nebular materials. That they moved at first in 
their partially free orbits in spiral paths sunward under the 
resistance of the residual, more attenuated, or less refractory 
nebular matter that remained by its elasticity above the 
denser part of the nebula nearer the sun's surface. That upon 
further shrinking of the sun's nebula and fall of temperature 
in a more regular manner, the earth's nebular zone was first 
formed as a mass extending in volume much beyond the 
moon's orbit under conditions already discussed. That this 
condensation, which finally formed the earth, was set in 
rotation consistent with its mode of formation due to the 



difference between the linear velocities of its outer and inner 
parts, which were moving at equal angular velocity, as before 
discussed ( 146). 

227, That under the above-stated conditions, the nebulous 
earth, being of much larger mass and condensing much later 
than the outer small planetoids, remained gaseous for a long 
time after they had condensed. That it became a liquid 
globe surrounded by a voluminous nebulous atmosphere at 
the time that the intra-Mars planetoids represented in minia- 
ture cold bodies similar to the present earth. That these 
planetoids after cooling possessed dense metallic centres, 
probably still highly heated, surrounded by oxidized metals 
and haloids, with water or, more probably, ice and air upon 
their surfaces. That the planetoids moving under the resist- 
ance of the still attenuated nebula surrounding the sun with 
their small excess of angular velocity, drifted slowly as regards 
inward motions sunward into the nebula of the new-forming 


globe until they were drawn to its surface and became finally 
incorporated about the perihelion positions of their orbits. 
The effects of the percussion again developing great heat, but 
not sufficient to convert the perfectly cooled planetoid pro- 
jected upon the earth's surface, if this was of large mass 
again into a liquid state, or to bring the earth and the 
planetoid back to the form of a single nebulous globe. 

228. Such systems of collisions as inferred above at every 
union of a planetoid, if this happened to be of large mass, 
might be considered to form at the time of contact a close 
binary system. The parts of the planetoid in contact with 
the earth would become highly heated by the effect of the 
collision, and sinking with its lighter coating into the liquid 
central globe of the earth would produce convection currents 
during its immergence, bringing the lighter liquefied surface 
matter into equilibrium of the mean gravitation system. In 
this process the lighter oxidized matter from the former buried 
surface of the injected planetoids being slowly floated up to 


the surface of the earth would remain projecting beyond its 
mean spherical surface, so as to form a future land-area. 

The extruded oxidized mineral matter of the planetoid 
would carry with it part of the metallic matter that forms the 
interior of the earth. Thus, through the friction of the opera- 
tion and cohesion, such dense matters as gold, platinum, and 
other metals formerly in the core of the planetoid would he 
placed much above their mean gravitation position on the 
earth's surface. 

229. The condition of our planet after the period of the 
last collision of a planetoid of any considerable mass would 
be most probably represented by an irregular spheroid sur- 
rounded by a nebulous atmosphere, still in a highly heated 
state from the temperature due to its original mode of forma- 
tion and from that developed by the collisions of the planetoids 
from which certain factors of its land-areas were possibly 
derived. We may consider a particular case. 

230. Projection of a large intra-Mars Planetoid upon the 
Earth. We will assume for the formation of a continent for 
a special case that a cold planetoid came into collision with the 
earth at the position of South America, which was of a mass 
greater than this continent, measured above the level of the 
ocean. That the planetoid moved in a spiral path under the 
resistance of the earth-nebula in the same direction as the 
earth's rotation, so that it reached the earth with what we 
may term a very moderate percussion. That the earth at the 
time was a liquid metallic globe, covered only by a thin 
coating of oxidized matter. The effect of the concussion 
would be that the surface of the cold planetoid would be 
raised to a white heat ; its water or ice would evaporate into 
the nebulous matter about the earth ; its oxidized surface 
would sink at once into the liquid earth-globe and become 
liquefied itself. This lighter liquefied viscous matter of the 
exterior surface of the injected planetoid, composed mostly 
of silicates as in the earth-system, would be floated up slowly 


about the periphery of the intruded mass, and the solid centre 
of the planetoid would gradually sink into the globe until it 
reached a position of gravitation-equilibrium and became a 
part of the interior metallic density-system of the globe. This 
could not occur, however, without much confusion and violent 
plutonic action within the central matter during the time that 
the lighter oxidized former surface-matter of the planetoid was 
reaching the permanent surface of the globe by convection 
currents. These currents would form special drifts and 
bring much of the metallic matter to the surface by tearing 
it away in collision, as it is a property of semi-fluid siliceous 
matter to bite into a metallic surface, as we find in the 
process of enamelling. 

231. The final result of such a motive system as here 
presented would be the projection of a mass of vitreous rock 
above the earth's surface, entirely added to the earth's super- 
ficial system, forming a hollow plane, produced by the excess 
siliceous material floated up at the periphery of the included 
planetoid. The lowest part of the surface of the earth 
covered by the projected planetoid would be raised above 
the mean gravitation-plane of the earth's surface, the entire 
upheaved rocks being in equilibrium-excess of mass of the 
volume of the vitreous rocks upon the surface of the partially 
included planetoid. The effects of the inclusion of a planetoid 
upon the early globe as herein described may be better 
shown by a diagram. Let fig. 22 A be the metallic core of 
the planetoid; 6Z/the surrounding tertiary matter; cc'the 
surface of the globe ; d d' d" the mass in section of the 
tertiary matter finally resting above the surface of the globe ; 
1 1' and t" t" 1 will show the lines of greatest mass that will 
rest above the surface. The tertiary mass in segment of arc 
z z' in a heated viscous state would float round the denser 
core and finally appear upon the earth at d and d'. The 
extruded tertiary matter would generally finally rest in 
gravitation-equilibrium with a part of its mass sunk in the 



matrix matter of the globe, and being vitreous it would also 
tend to run down upon the surface, but in its early extrusion, 
being at the same time subject to the radiation of heat when 
projected much beyond the earth's surface, it would partially 
cool down and finally take a mean gravitation position in 
relation to the dense metallic core of the earth with con- 
siderable projection therefrom. 

Fig. 22. 

232. As the planetoid may be assumed to strike the globe 
near its equator, it would have more of its matter included 
upon the equatorial than the polar side of contact. The 
plutonic mass would therefore tend to extend poleward, 
diminishing its volume so as to leave the continent due to 
its projection of pear-shaped outline, with its broadest end 
towards the equator. 

233. The cold planetoid falling towards the earth-centre 
would leave its upper surface exposed for a period, and this 
shutting-offthe radiation from the highly-heated globe beneath 
would cause a rapid deposition by condensation of the super- 


imposed nebula, so that it would acquire a heavy surface- 
stratum of new deposition beyond that of the planetoid's mass. 

234. In the above we have taken an extreme case, The 
special condition of the fall of small bodies of meteoric matter 
could not have produced distinct features upon the earth's 
surface. If meteoric bodies fell in swarms, the effects would 
be the same as those just considered for single planetoids, 
but probably with more dispersion upon the earth's surface. 
The general detached fall of single small meteorites could only 
produce a certain amount of incidental inclusion of that which 
appears to be foreign matter to the system of stratification 
wherein it fell, which is recognized by geologists. 

235. It cannot be assumed that the foregoing conditions of 
projection of large planetoids could remain permanent, being 
since subject to long periods of atmospheric denudation and 
all the after- conditions of our globe, of which we have full 
knowledge by geological evidences that remain in volcanic and 
stratified rocks ; but upon the whole the initial setting-out of 
the globe in local land-areas, if it occurred by the inclusion of 
planetoid matter in the manner proposed, could never be 
eradicated. The forms of ancient continents have no doubt 
been modified in time by atmospheric conditions, and by the 
effects of oceanic currents, through the direction given by 
solar impulse, causing wear in their projecting parts and 
deposition of matter in quiescent contiguous water, owing to 
which the great ocean-basins at the present time approach 
circular areas of circulation of free oceanic surface, which I 
have considered in my work on fluids * (fig. 22) . Under these 
conditions the central continental matter may in some cases be 
undisturbed. The general mass of vitreous matter, localized 
in great depth of rock by such projections as suggested, could 
never be entirely distributed by forces known to us as having 
been active upon them. 

* < Fluids,' 1881, pp. 354, 377. 

[ 169 ] 



236. Earth-formation under Nebular Conditions only. This 
was proposed ( 204) as the most important factor. The 
proposition of land-formation from discrete matter given in 
the previous chapter may be considered as purely incidental, 
affecting certain land-areas and occurring at indefinite 
periods. On the other hand, the conditions of nebular con- 
densation to form the earth must have been constant during 
the radiation of its initial heat at the time that such nebula 
could form a dense atmosphere around the earth. It will be 
therefore convenient for distinction and simplicity to consider 
this nebular condition separately as regards its entire effects, 
neglecting for the time the conditions which have been already 
discussed of incidental incorporation of solid planetary matter 
with the earth. This will not change the discrete conditions 
of projection of matter before proposed, which must have 
entered into composition with the nebular conditions now to 
be considered, as it is in this combination only that we can 
have the entire evidences of the actual conditions satisfac- 
torily explained. 

237. We have already considered many modifying con- 
ditions that may have been active during the formation of 
our earth, particularly in the effects of the state and condition 
of the sun at different periods. Yet we must assume that 
there was a general or uniform system of condensation of 


the nebula which formed the earth due to radiation of its heat 
only into space, upon which the varying outward conditions 
of the state of the sun would he superimposed. Under such 
purely nebular conditions we will assume for the present 
that the sun's condensation may be taken as a constant, and 
that the temperature of the earth's nebula was decreasing at 
a uniform rate. The more special condition of variation 
will be considered in the following chapters. The factors of 
purely nebular conditions will be most conveniently taken 
under separate headings. 

I. Period of dissociation of elements onward to the time 

of the early condensations which were adapted to 
finally produce the globe in a liquid form. 

II. Period of condensation of volatile metals, association 

of oxygen and the halogens with metals and metalloids 
to form the earth's crust. 

III. Period of deposition of water. 

IV. Period of deposition of snow and of ice formations. 
This last period is very important in completing the causes 

of changes that have passed; although in this we go beyond 
purely nebular conditions, still the mode of action is con- 
tinuous. This subject may be better considered in a separate 

These periods, although it is proposed to take them 
separately, run necessarily the one into the next by inter- 
mediate stages, as the separate systems defined depend upon 
the effects of time which is continuous. 

238. I. Period of Condensation of Highly Refractory 
Matter. At the commencement of the first period suggested 
above, we may assume that, immediately after the separation 
of the earth's nebular ring from the sun's nebular system, 
the heat at this time may have been sufficient for the disso- 
ciation, at least so far as a gaseous or vapourous condition, 
of that which was to become terrestrial matter. At such a 
time, if the nebular ring were denser in any part or if it 


were broken by any disrupting cause, such as the intrusion 
of a comet, a local density system would be formed in some 
part, which would then become the nucleus of our earth. 
The ring of nebula after its detachment from the sun is 
assumed to be in equilibrium in its mean orbit, so that all 
parts of its matter would drift slowly towards any denser 
parts by consecutive attraction of its near parts. In such a 
system if the nucleus was formed by heavier heated matter than 
the condensation of the exterior nebulous matter, this would 
drift thereto and approach the centre in convection currents. 
These currents would be constantly active in proportion to 
the difference between the density of the central system and 
that of the surrounding vapours or gas to cause the speci- 
fically heavier vapours, as, for instance, those of the heavier 
metals, to move through the minor resistances of the lighter 
gases towards the centre of the svstem. At the same time 

o / 

light elastic gases would be buoyed up to the extreme outer 
circumference of the system. This is exactly the case at the 
present time with the exterior of the chromosphere of our 
sun, which we find surrounded by hydrogen, the lightest 
known element, with denser gaseous and vapourous matters 
placed concentrically beneath *. 

239. Immediately following the first condition of an exten- 
sive nebulous or gaseous atmosphere surrounding the new 
globe-system forming from the continued process of radiation 
of heat, there must necessarily come a period of con- 
densation of the nebulous matter to liquid or solid matter. 
Whenever this period arrived it would be quite indifferent 
in what part of the nebulous ring system about its exterior 
surface the condensation occurred. The condensed matter 
would, by its superior specific density, immediately pro- 
ceed to pass through the lighter gaseous parts towards the 
centre of gravity of the system, even although the system 

* Lockyer, ' Studies in Spectrum Analysis,' p. 147. 


might remain almost entirely nebulous, just as rain passes 
through the atmosphere. Thus, if we imagine the platinum 
group of metals to be widely diffused in the nebulous state, 
these would not only tend to take a central position by 
gravity, but in all possible condensation through exterior 
radiation they would flow towards the centre. Particu- 
larly in this case, as such matter as platinum would have 
little affinity to unite with the oxygen or halogens present 
through which it might pass. Further, this metal would 
condense at a temperature so high that the more general 
matter prevailing would remain gaseous. Like conditions to 
those just explained for platinum would hold also with all 
other dense refractory elements, causing constant exchanges 
of relative gravitative positions throughout the long period of 
condensation until the densest and possibly the least oxidable 
of the original nebular matter formed a considerable globe. 

240. Whatever may have been the composition of the 
early condensations or primitive liquid matrix which formed 
the early earth, whether this was metallic or mineral, in the 
ordinary sense of the term, one thing is certain, that the 
present crust as we know it is very largely composed of 
metallic and siliceous oxides. It is therefore reasonable to 
suppose that oxygen, being lighter in its pure or gaseous state 
than the average mass of the early nebulous globe, remained 
in the early highly heated nebula an exterior element, and 
that compounds or oxides derived therefrom were deposited 
at a later period than the refractory metals. 

Under the above conditions, assuming that the denser, 
more refractory matters continually condensed about the 
central system or globe, this condensation might occur even 
while the surface maintained an almost incandescent tempera- 
ture ; so that possibly at a period when the earth measured not 
less than within 400 miles of its present diameter, it was still 
a liquid globe at a white heat comparable in temperature at its 
surface with that of the lavas which issue at the present time 


from our volcanoes. At this period the globe would necessarily 
be surrounded by an atmosphere which contained as a part 
of its earliest gaseous density arrangement nearly al| the 
oxygen, chlorine, and hydrogen of our system, together with 
all other elements which remain gaseous at the high tempera- 
ture now considered, except only such part of them as might 
be condensed or combined near the surface of the globe 
by the enormous nebulous atmospheric pressure then prevail- 
ing. The globe, unless disturbed by collisions with exterior 
matter, would be at this time an incandescent smooth uniform 
liquid spheroid, without any projections upon its surface. 

241. From the mean density of the earth, 5*6, it is highly 
probable, as before proposed, that the central matrix is entirely 
metallic. Under this condition, however distinctly the con- 
densations might form at the earliest stage an increasing 
density system towards the centre, the tendency of metals 
to form alloys would materially modify this during the 
after condensation of the lighter metals. Further, the heat- 
conducting powers of the metals would tend to keep the 
globe at a uniform temperature to a depth below the point 
where matter could experience the effects of radiation. 

There is no reason to suppose that these conditions have 
been materally modified as regards the greater part of the mass 
of the earth up to the present time, as there could not have 
been great loss from radiation during the time that the earth 
was surrounded by a dense nebulous coating which was 
giving out its heat on continuous condensation, nor after the 
oxidized coating had condensed to form our present highly 
non-conducting surface, as this would give out great heat 
upon its oxidation. Therefore the probability is that the 
loss of central heat was and is largely operative only in con- 
densing or thickening the surface coating, under conditions 
which will be discussed further on. 

242. At the period when the earth was wholly liquid the 
outward diameter of the sun or his photosphere probably 


extended much beyond the orbit of Venus. At this time 
the earth was possibly outwardly a self-luminous globe partly 
obscured and surrounded by a moon-ring or nebulous moon. 
Possibly also the planet Jupiter was not wholly condensed, 
but appeared as a large heat-giving nebulous body dispersing 
a small part of this heat to the earth. Under these conditions 
the earth's equator by reciprocal radiation-of-heat exchanges 
with the sun, and in a less degree with the other planets, main- 
tained nearly its initial temperature for the time. The polar 
regions being open to space would be radiating their initial 
heat more rapidly than the equatorial regions, thereby re- 
presenting the area of cooling surface to which condensing 
nebulous matter near the earth's surface would drift in aerial 
currents, a condition which, as regards the vapour of water, 
has remained to a certain extent permanent. 

243. II. Period of Condensation of Volatile Metals, 
Association of Oxygen and the Halogens with Metals and 
Metalloids. We may now consider the conditions of a period 
when the sun's photosphere had shrunk to nearly the 
orbit of Venus, and heat exchanges within the planetary 
plane had grown much less active ; when the polar regions 
through radiation of their initial heat may have been reduced 
to a red heat and assumed a viscid consistence. Under these 
circumstances, by the continuous radiation of heat from 
the exterior nebulous surface or atmosphere, and the cooling 
by radiation of initial heat from the surface of the globe, 
volatile metals would be deposited and oxygen and the 
halogens would approach more nearly to the earth in its 
nebulous atmosphere, by which means their association with 
metallic vapours near the surface would be possible and 
might even be rapid. At this period, under the great 
nebular atmospheric pressure then present, the new-formed 
oxides would fall as rain over the cooler polar regions, where 
the surface heat could not support the vapourous state ; and 
we may assume that at their great heat they would remain 


in a viscous state, and therefore would flow outwards from 
the cooler polar regions towards the equator to gravitative 
equilibrium, combination with the central matrix being no 
longer possible, as in the early conditions with prevalent 
metallic surface matter. 

244. In continuity of what is here taken to be the same 
period we come upon conditions most important to geology, 
namely when it became possible as a natural result of the cooling 
for solid oxides, when they formed tertiary matter, to rest upon 
the surface of the globe. This occurred at first possibly as a 
kind of oxidized scum, such as we find floating in the ordinary 
melting of metals exposed to the air, and no doubt only over 
the colder area of the poles as just stated. The presence of 
such a scum acting immediately as a non-conductor of heat 
would shut off a part of the initial radiation from the surface 
of the central matrix of the earth, which previously kept the 
gaseous matter above it from condensation at nearly its 
critical point. This loss of the support of radiated heat from 
the earth would cause the condensation of vapourous and 
gaseous matter to appear as dense clouds and finally to pre- 
cipitate the condensed oxides, of which the clouds were formed, 
upon the cooler polar areas of the earth. Under these con- 
ditions, as soon as a cooler primitive scum was formed, this 
would be immediately covered by further nebulous deposition 
of oxidized matter. 

245. As the polar regions of the earth cooled down through 
open radiation the early basic matrix would no longer main- 
tain sufficient surface-heat to retain a liquid surface, and the 
early siliceous oxides would become by loss of heat stiffly 
viscous. After this period, the same oxides that fell at great 
heat in a liquid or viscous state might fall over the cooler 
areas as a kind of tertiary mineral snow, and in this way 
cover up the primitive viscous scum, possibly even to very 
great elevation. At the same time this would depress the 
supporting liquid and the viscous surface beneath, causing 


this surface-matter to move outwards from the poles to equi- 
librium of gravitation by displacement. In this manner 
gravity would spread out or carry the newly-formed oxides to 
great distances from and around the poles. 

246. In the above scheme it is assumed that oxygen and 
other elements of low specific gravity at a high temperature 
could not reach the earth for combination with denser metallic 
matter at any early period of condensation. But if we suppose 
there were certain combinations which would be precipitated 
at any given temperature from the nebula, even before the 
complete metallic matrix was formed, their specific gravities 
would still maintain them as a scum, and the after con- 
ditions, as already described, would not be materially modified. 

247. Long after the deposition of the more refractory oxides 
considered above, we should naturally have the deposition of 
the more volatile chlorides, and in a less degree the fluorides 
and iodides, and generally we should have the most volatile 
matter of the solid earth longest in suspension in its primi- 
tive nebular atmosphere. At this point it therefore becomes 
important to consider what bodies in combination may have 
remained gaseous at a moderate temperature capable of finally 
contributing to the matter of the earth's crust, which we 
may assume, from the presence of their elements in the 
early rocks, were possibly largely diffused previously to the 
formation of our present atmosphere. Of these gases as 
most important we may take : 

Aluminium Aluminic chloride, A1 2 C1 6 . 

Iron ... Ferric chloride, Fe 2 Cl 6 . 

Or, descending to our present atmospheric temperature, we 

r Silicic hydride, SiH 4 . 

Silica < chloride, SiC 4 . 

(. fluoride, SiF 4 . 


r Carbonic oxide, CO. 
Carbon -3 Carbonic dioxide, C0 2 . 

C Carburetted hydrogen, CH 4 . 
Sulphur . ( Sulphurous anhydride, S0 2 . 

(.Sulphuretted hydrogen, H 2 S. 

248. To which we may add the vapour of water. In some 
cases one of these gases would decompose the other and cause 
deposition. There are other compound gases, less important 
in a geological sense, that would remain constantly present 
at a somewhat higher temperature. There is an omission 
in this volatile series of two very prevalent surface materials, 
calcium and magnesium, the chlorides of which are gaseous 
only at high temperatures. But the oxides of these bodies 
are so extremely light, that assuming they condensed with the 
heavier siliceous oxides they would fall very gently through 
a dense resisting medium, such as the nebulous atmosphere at 
this period is considered to be. Further, from these oxides 
being very light and refractory in a finely-divided dry state, 
they would remain upon the surface and drift about in the 
dense aerial currents then prevailing, and rest finally in 
positions above the heavier siliceous and denser metallic com- 
pounds, which would be for the greater part condensed at 
the time. Under any conditions, by their specific gravity 
they would take finally a surface-position upon the primitive 
scum. The condensation of the above-described compounds 
with others present, possibly formed a considerable part of the 
superficial crust covering the earlier deposited oxides; but 
at this time we commence to have the interference of water 
condensed in large volumes, which was instrumental in many 
changes to be considered further on. 

249. Formation of Continents. We may now consider the 
effects of the above-described formations generally upon the 
earth's crust as influencing the formation of continents, 
under conditions of nebular condensation only. We may 



take it that at the period when oxygen and the halogens 
entered into association with the prevalent basic metals, their 
oxides were first deposited over the cooler polar areas upon 
principles just discussed, that all the surface of the globe, 
except the polar regions which were most open to radiation, 
was in a highly heated liquid state, and the whole substratum 
of the polar area in this state also. Under these conditions, 
as the oxides condensed about the cooler areas of the poles 
this lighter oxidized matter, its lower part being in a plastic 
viscous state loaded up as it might be by condensation at the 
poles to great height, would have through the influence of its 
gravity a constant tendency to be pressed outwards from these 
regions, so as to float upon the surrounding liquid metallic 
matrix into lower latitudes. At the same time, by natural 
affinity these oxides would cohere together, more particularly 
at the base of the mass, where the internal heat and pressure 
would maintain them in a liquid viscous state. Thus as the 
compound oxides accumulated in mass they must of necessity 
have overflown or be thrust outwards from the cooler polar 
regions from their own internal gravitation-pressure, and in 
this manner form coherent projections, taking the lines of 
least resistance in moving upon the liquid or semi-liquid 
metallic matrix constantly towards the more highly-heated 
regions of the equator to restore gravitation-equilibrium. This 
thrusting-out would occur exactly in proportion to the accu- 
mulation of the lighter condensed oxidized matter from its 
increased pressure through constant deposition at the cooler 
area of the poles. 

250. The form which the above- described coherent pro- 
jections of oxidized matter would take would be that of long 
tears, precisely the same as that of the boundaries of a melted 
metal or other cohesive liquid poured out continuously in bulk 
upon the centre of an extensive heated level slab. This form 
would, however, be modified to a certain extent by the re- 
melting of part of its extreme prolongation, particularly at the 


thinner bordering edges in approaching the more highly 
heated, that is the less cooled, equator. These prolongations 
from the polar regions would therefore take the final forms 
of pointed tongues, maintaining a wide base upon the polar 
regions and diminishing gradually as they approached the 
still almost incandescent tropics, where heat would be partially 
held from radiation by the large predominant sun. 

251. Under the above conditions there would be also present 
a constant tendency to widen the base of the prolongations 
above described. Firstly by the extension of the solid surface 
of the polar regions through continued cooling and the 
increase of the amount of support to the elevation of oxidized 
matter through submergence of the lower parts. Secondly, 
by the deposition of a greater amount of nebulous matter, 
which would increase proportionally to the surface shading 
from the influence of the radiation of the earth's initially- 
heated matrix, so that from both these causes a constantly 
wider area of base would be added to the floating prolongations 
as they were pressed outwards from the polar regions, making 
the entire projections of conical form. At the close of the 
period above depicted, so far as condensing nebular conditions 
were concerned, the whole of the new-formed land-surface, 
derived from the deposited oxides, would consist of pointed 
areas of projection extending towards and over the equator, 
proceeding from an elevated base spread over a wide region 
around the poles. The new land in lower latitudes would 
appear as a floating mass, possibly of a dull red heat, upon 
a liquid matrix of much greater density and more highly 
heated than itself. 

252. We may now consider a period when the whole 
surface of the globe proceeding from the poles began to set 
to a viscid resisting surface as an effect of the radiation of its 
initial heat into space, and when the new land-surface in 
pointed continental areas had become superficially rigid. At 
such a period the central areas of the new land, being pro- 



tected by a non-conducting coating from excess of radiation 
from the still highly-heated matrix beneath, would possibly 
retain a semi-viscid condition at a small depth. We still as- 
sume that mineral matter is being deposited, although more 
slowly, from space, especially at the cooler areas of the poles 
now assumed to be much below red heat. Then, however 
much mineral matter should be piled up at the poles of the 
earth, it would be more and more resisted at the borders 
of the now viscid surface of the matrix surrounding the 
new continents. Under such conditions possibly the edges 
of the new-formed continents would be pressed up and 
contorted by the resistance to further projection and have 
their borders thrown up much beyond the average surface, 
such edges being continually cooled and consolidated by 
their exposure to radiation at the greater heights. New 
smaller tongues of projection might break out here and there 
at any points of least resistance for the escape of the internal 
pressure of the semi-liquid oxides flowing to equilibrium. 

253. At a later period, when the floating movement of the 
continental areas became quite impossible by the resistance 
offered by a considerable depth of the cooled surface of the 
matrix and oxidized magma above, and the various matters of 
the nebulous system were mostly deposited, the more heated 
interiors of continents then formed where the polar outflow 
was most continuous, would contract and sink down in cooling, 
and leave the new land-areas, although of great altitude, of 
basin-like section. 

254. Distribution of Land-areas. To summarize the con- 
ditions here proposed : the continents may have been fully 
delineated before water yet filled the oceans, these appearing 
as pointed land-systems proceeding equally from each of the 
polar areas ; but if we consider the actual condition we find 
we have very little land in the southern hemisphere and much 
in the northern. There must, therefore, have been, even at 
this early period, some modifying conditions to those now 


generally proposed to account for this. What such conditions 
were has been already suggested upon the precipitation -theory 
in the last chapter. They might possibly also, perhaps less 
logically, be suggested upon the nebular theory simply. 

255. If at the period when the earth was still an incan- 
descent globe, we assume that radiation of heat from any 
cause might have been greater in the northern than in the 
southern polar regions either from there being cooler space 
towards the northern pole or from deficiency of heat- radiation 
from distant celestial bodies, or from excess of winter eccen- 
tricity of orbit at the time, or other causes sufficient to chill 
this pole first, then the first accumulation of oxidized 
deposits would immediately shut off radiation of initial heat 
from the early matrix of the globe into the superimposed 
nebulous matter and cause its precipitation at this pole only ; 
and as the area thus shaded from internal radiation by the 
first early deposits would constantly increase in dimensions 
proportionally to the depositions, the pole first chilled by 
any cause would become in increasing ratio the greatest area 
of deposition. In this manner the land-forming oxides might 
be deposited at the north pole in quite sufficient excess to 
account for the difference we find at present between the 
amount of land-surface in the northern and that in the 
southern hemisphere, or be even sufficient to disturb the 
position of the centre of gravity of the earth's mean volume. 
So that at the present time the superficial subaqueous region 
of the south pole may be of much denser matter than that of 
the north where the early oxides were most deposited, which 
condition must continue permanently. 

256. If we now take cognizance of the present form and 
distribution of land-surface and oceanic basins, and make 
allowance for all reasonable changes, volcanic and other, 
that the immensity of time must have brought about be- 
tween the periods already discussed and the present, we may 
ask, Is the distribution of land entirely consistent with the 


theory proposed upon nebular conditions only ? This would 
be very difficult to answer in the affirmative, but we may find 
that in certain points there is agreement, which may possibly 
be all we may expect after so great a lapse of time, and with 
other factors of formation that have been already proposed. 

257. By the nebular system here suggested, we should 
have a mass of land at the north pole from early condensa- 
tions, and a less mass at the south pole from later condensation. 
Both of these land-areas would possess pointed prolongations 
extending outwards to or possibly even over the equator. 
Of the actual land-areas which may be considered as consistent 
with the principles discussed in this chapter, we may take the 
prolongations of land in North America with arctic base, the 
lesser prolongation of Greenland, the prolongation of the 
Europseo-Asiatic continent to Mocha, that of the peninsula of 
India within the Ganges, and that of Siam, extending at one 
time possibly to Sumatra. We now find the most important 
areas of S. America, Africa, and Australia disconnected from 
polar areas. But as regards Africa, if we take a wider view, 
this may be considered as a prolongation of the Europseo- 
Asiatic area, leaving then only the difficulties of accounting 
for the division of the narrow Mediterranean and Ked Seas. 
In the same manner, Australia having the broadest base to the 
south, we may imagine it to have been once connected with the 
antarctic polar system. At this period its area would extend 
in pointed form, including and possibly extending beyond 
New Guinea. The principles discussed leave South America 
an entirely detached continent, which, considering its theo- 
retical form upon the nebular principles just proposed, would 
indicate that it must belong to the northern area, but we 
now find it in the southern. Therefore, this may possibly be 
better accounted for upon the discrete conditions previously 

258. It is not probable that the early conditions of earth 


formation could be fully recognized, as very material changes 
must have occurred since. The anomalies to the theories 
proposed of the detachment of S.America, Africa, and Australia 
might possibly be met in a certain degree by supposing that 
these cohesion systems were detached from the area of their 
polar condensations and drifted when the earth's metallic 
matrix was still superficially liquid into their present position. 
This might have been caused by outward pressure from the 
poles, of matter which condensed at a certain period, but 
afterwards dissolved in the highly heated waters, deposited 
in great volume and under great pressure, which will be 
presently discussed. Taking all these matters into consider- 
ation, however, it is not probable that nebular conditions alone 
ruled, as there were no doubt also discrete condensations 
about and without the nebular zone or ring, which in time, 
through crossing orbits with the earth, came into collision with 
it and materially modified the land-areas, possibly in the 
manner discussed in the last chapter. 

259. Following the above-described hypothetical conditions, 
under which the basic superficial system of the globe may have 
been formed, we have for the consideration of the present 
continental forms to allow for the influence of forces acting- 
onward to the present period, to which we must attribute great 
modifications. Among such influences we have erosion and 
after deposition on coasts, vulcanicity in its widest extent, and 
the wear of oceanic currents, forming altogether constant 
elements of modification with others to be discussed. 

260. Whatever may have been the special or local con- 
ditions of early continental formation, it is very probable 
that at the close of the long period of deposition of oxide and 
halogen compounds, or what we may term the dry period of 
deposition, the land-areas were generally clearly defined for 
all future time. If these oxides as non-conductors stopped 
the radiation of central heat and at the same time radiated 


their initial heat from the surface, the deposits would be 
piled up over such areas, and by this loading press down 
the lower surface of the land to gravitation- equilibrium, so 
that the land of the earth may be represented as consisting 
of masses of oxidized matter floating upon a lower denser 
liquid metallic substratum. 

261. From the immense volume of oxides known to exist 
upon the surface of the globe, by the mode of precipitation 
here proposed the land-areas may have attained great height, 
possibly locally of ten to twenty miles. These areas, ob- 
structive to radiation of internal heat, would advance from 
the poles, so that afterwards deposition of oxides or halogens 
over such land-areas would become general, and upon points 
of elevation as great as over polar areas, in the same way 
that snow falls on mountains at the present time. In such 
new-formed depositions, generally pressing from their centres 
outwards to gravitation-equilibrium, acting with deposition of 
water, we have possibly the entire factors of the great gneiss- 
forming age, -vestiges of which remain intact with all the great 
displacements and contortions impressed upon it which are 
still evident at the present time. This will be discussed 
further on. The depositions about the tropical land-regions 
which may have been produced by causes discussed in the 
last chapter, would act in conjunction with the condensations 
here considered and locally magnify their effects. 

262. III. Period of Formation and Deposition of Water. It 
has been calculated that the whole of the water upon the globe, 
if equally distributed, would be about two miles in depth. If 
this, at the temperature then prevailing, formed an atmosphere 
about the globe it would produce a pressure of about 400 
atmospheres, or 7000 Ibs. per square foot of surface. It is 
unnecessary to say that this could scarcely occur at a time 
when the earth's surface was even at a red heat, for at this 
pressure water would possibly remain liquid even at a white 
heat. Further, by the classical experiment of Cagniard de 


Latour *, water at a temperature of 412 C. was found to 
dissolve glass. Further experiments have shown that siliceous 
rocks could not only be dissolved in water by heat and pres- 
sure in the presence of alkalies, but be recomposed to form 
quartz, felspar, and other minerals f. At the heat and pressure 
then present, water would therefore combine with mineral 
matter, which would be crystallized and become solid rock as 
the temperature afterwards diminished. We must therefore 
suppose no distinct definable line for the formation of water 
on the globe, but merely make a broad assumption that it 
probably formed after the deposition of the mass of the most 
refractory oxides which could remain gaseous at a very high 
temperature only. Further, there is no doubt that the 
energetic action of the early highly-heated water would be 
greatly increased by the presence of corrosive mineral acids 
within it, which would be in a vapourous state at the same 

263. We may imagine at the first great deposition of water 
which occurred at the cooler area of the polar regions, that 
by its great solvent power when at great pressure it would 
at once reduce and dissolve many of the oxides, and more 
particularly the siliceous ones. That the streams produced 
would channel out and disintegrate the solid elevated siliceous 
deposits then covering the polar areas, forming great rivers, and 
finally, by the heavy constant highly-heated rains, divide the 
land up into detached parts or islands, carrying the dissolved 
matter into the deep equatorial oceanic basins to gravitation- 
equilibrium. That as the tropical areas by the conditions of 
mutual heat-exchanges would be much more highly heated 
than the poles, this water would again, for a large part, 
evaporate over the tropics and leave behind the mineral 
matters crystallized at the bottom of the ocean. In this 

* ' Annales de Chimie/ IE. xxi. & xxii. 
t Qeol. Soc. Trans, xv. p. 103. 


manner the ocean bottoms would be formed from the debris 
of the early polar condensations, and the polar areas would 
become channelled and in time become largely oceanic by 
this constant denudation and chemical action of the constant 
drenching of the highly heated polar rains. 

264. If we may assume that the process suggested above 
continued until the sea reached to within half a mile or 
so of its present level, the condensation of vapour might 
then have become possible over the now much cooled lower 
latitudes, and the elevated tropical lands possibly began to 
receive a copious rainfall. It is throughout this period 
until the ocean became of nearly its present level that we 
may look for the greatest aqueous deposits upon all lands 
in inland seas and shores together with a large deposit over 
the bottoms of the oceans. The deposits as they were washed 
from the coasts and inland would be left by gravity piled up 
against the borders of the continents, and, generally by the 
action of currents carrying mineral matter, irregularities and 
depressions upon and about the continents would be filled up 
or smoothed off *. 

265. Of the gaseous matters previously considered as 
possibly remaining longest in the atmosphere and incidentally 
under decomposition forming surface matter ( 247), the 
most important are the chlorides, and these are not found 
largely in the earlier rocks, but in the ocean. The reason 
for this is probably that they were decomposed in the presence 
of oxide of sodium ; and as sodium does not form a gaseous 
compound with chlorine at ordinary temperatures, it would 
be at first precipitated on the earth from the nebulous en- 
velope as one of the later lighter oxides, and this oxide after- 
wards in the presence of water would decompose the chlorides 
present, and still remaining in solution would carry the chlorine 
to the ocean as salt, leaving its oxygen reunited with elements 

* ' Fluids,' by the Author, p. 373. 


present which were not saturated to the highest or most per- 
manent state of oxidation. 

266. If we take the termination of the aqueous period to 
be that at which ice was first possible of formation at the 
poles, this would be the time of greatest general elevation 
of the ocean ; for although we may suppose that a much 
greater volume of water was held in vapour in the atmosphere 
by the heat conditions present, still this would not nearly 
equal the enormous amount of ice at present elevated at the 
poles above the oceanic level which in the case assumed 
would form part of the ocean. Dr. Croll * has estimated 
that the melting of the antarctic ice-cap alone, considering 
this as only equal to one mile in thickness, would raise 
the general oceanic surface 200 feet. A mile is probably 
much under the truth for the elevation of the interior of this 
area; but if we add to this the arctic ice above sea-level, and 
that of all elevated ice-clad regions, the general level of the 
ocean would be raised possibly more than 400 feet from its 
present surface, which, added to the present sea-level, might 
indicate approximately the level of the ocean at a period 
before ice was formed. Further, it may be presumed that 
the constant deposition of water must have naturally reduced 
the amount of land, so that the lower lands of the present 
surface were entirely submerged. 

267. This aqueous period, as will be hereafter considered, 
possibly occurred in the palaeozoic age and also later in the 
miocene period, when a temperate climate extended to the 
north of Greenland. The general effect of these aqueous 
periods upon what was at the time, and what afterwards 
became, low continental land, was the heavy and almost con- 
tinuous deposition of the debris of originally deposited matter 
brought down by constant rainfall to the shallow coasts and 
inland seas. This deposit upon principles suggested, although 

* < Climate and Time/ p. 388. 


less in proportion as it was distant from land surface, must 
have been greatest towards the polar regions, partially filling 
the oceans in these regions and generally diminishing towards 
the equator. Therefore it is improbable that such enormous 
deposits as the northern silurians will be found in the tropical 
regions, their representatives being much thinner strata 
derived from the ddbris only of the elevated continental lands, 
produced entirely by causes already discussed. 

[ 189 ] 



268. Pre-glacial and Glacial Periods. We have now to 
consider the fourth period ( 237), when the residual nebular 
matter surrounding the earth consisted, as at present, of air, 
water, and carbonic anhydride. During this period, which 
extends to the present, the earth-surface, from the effects of 
the sun acting upon it, may be conceived to represent a heat- 
engine in which the tropical regions are evaporating the 
surface-water which is being simultaneously condensed about 
the poles. Under these conditions the probable geological 
effects of accumulation of snow about the poles will be now 

269. At what period ice could first accumulate about the 
poles would depend upon the amount of secular cooling of 
the earth's surface and the amount of heat received from the 
sun. Difference in the effective amount of sun-heat might 
be brought about either by diminution of his volume or by 
clouding effects which may have occurred through condensa- 
tion at any critical point of temperature of the solar nebula 
( 95). The diminution of the sun's disc would be constant 
and regular : the clouding effects would be exceptional and 
depend upon the chemical constitution of the nebula at the 
time. These exceptional conditions due to clouding will be 
considered hereafter in another chapter ; the regular or 
symmetrical condition due to the diminution of the sun's 


disc will now be suggested. These symmetrical conditions 
were discussed by me in a paper read before the Geologists' 
Association, March 2nd, 1882, and are now reproduced with 
some slight additions *. 

270. Ice, of which we have the greatest mass at present, 
under the uniform condition of condensation of the sun, 
probably formed when his apparent disc was not more than 
ten times his present diameter and when the mean surface- 
temperature of the earth did not at most exceed its present 
tropical temperature. At this period we had a much more 
aqueous atmosphere than at present, as it is the law that the 
quantity of vapour in the atmosphere when this is saturated 
increases in geometrical progression as the temperature in- 
creases in arithmetical. We had, therefore, probably at this 
period, considering the relative areas of the globe under 
tropical and temperate temperatures, a mean of about ten 
times the present amount of aqueous vapour in the entire 
atmosphere. At this relatively warm period as compared to 
the present, we have only to imagine that a polar area became 
sufficiently cooled through excess of radiation of initial heat 
from the earth for condensation of water to occur in the solid 
form of snow, and we have then the certainty of a continuous 
copious fall of snow over the same cold regions during the 
winter in the place of a former rainfall. 

271. If we take the process of the cooling of the earth as 
being quite gradual, omitting variations in the heat-giving 
power of the sun which may have been brought about under 
conditions already considered and other effects to be discussed 
in following chapters, we may suppose that the cooling of the 
earth sufficient for the deposition and formation of ice would 
at an early period produce very little geological change, 
if we omit consideration of all effects upon animal and vege- 
table life, which would be materially affected by frost. The 

* < Nature,' March 29, 1883 ; Proc. Geol. Assoc. vol. viii. p. 89. 


ice formed at this early period in the winter would be dissi- 
pated in the summer, uncovering the land-surface and leaving 
it at about the same level, except for a small amount of 

272. After the early period defined above, the immediate 
effect of the further cooling of the earth from any cause, 
astronomical or secular, would be the greater deposition of 
snow in high latitudes, which as it constantly accumulated 
in mass would slowly bring about the proportionate lowering 
of the oceanic surface upon the entire globe from abstraction 
of the surface-water. Under these conditions the littoral areas 
formerly submerged in shallow water would be gradually but 
slowly uncovered, until in the course of time the present 
extent of land-surface appeared. 

273. If we assume upon the symmetrical conditions pro- 
posed, jbhat the entirely aqueous epoch closed with the miocene 
period, when ice of the present system, due to decrement of 
the sun's volume and other causes, probably began to cover 
the poles in winter and melt in the summer, this would 
evidently be the period of the greatest extent of oceanic 
surface, for not only would the waters of the ocean be dis- 
tributed over its surfaee to gravitative equilibrium, but the 
land would have become largely levelled down by the heavy 
rains of the earlier period, when the atmosphere was more 
highly charged with vapour. When the elevation of ice at 
the poles after this period slowly abstracted a much greater 
part of the waters of the ocean, much of the shallow muddy 
shores must have become soil adapted to vegetable growth 
over the temperate regions. 

274. The continuity of the oceanic depression upon the 
conditions just stated and contemporary circumpolar elevation 
by deposition of snow, as it changed the extent of land-areas, 
must have affected the land-resistances to the direction of the 
projection of oceanic currents, and with them the superimposed 
air-currents, and have caused local variations of temperature 


in polar and temperate regions by the direction given to these 
currents due to heat and expansion of air and water from 
the diurnal impulse of the sun *. 

275. The elevation of snow to great height by condensation 
at the poles would by its pressure upon the yielding mass of 
the globe, whose equilibrium could only exist in a form 
consistent with its gravitation and rotation as a spheroid 
of revolution, cause an excess of pressure upon polar areas, 
which would react upon the lower viscous superficial strata of 
the earth and cause extrusion of viscous rocks to the surface 
or the elevation of certain areas of the surface which offered 
the least resistance to the imposed internal pressure until 
approximate equilibrium of the rotative gravitation system of 
the globe was restored. This elevation would be brought 
about by the extrusion of felsitic or basaltic rocks to the 
surface, or slowly in the upheaval of extensive land-areas, or 
more violently in earthquakes and volcanoes, according to the 
state of resistance due to the density, flexibility, or previous 
faulting of the more or less yielding surface-rocks. 

276. It must be impressed that the necessity for the main- 
tenance of land-areas depends upon the elevation of rocks by 
plutonic forces. Degradation by atmospheric forces and tidal 
action is constant, so that unless the elevation is in excess of 
the depression in the ratio of this constant degradation, land- 
areas must constantly diminish by levelling-down, and, judging 
from the quantity of sediment known to remain at present 
in stratified rocks, all land-areas must therefore have been 
wasted away ages ago. 

277. The Distribution of Ice at the Poles. Under the con- 
ditions discussed 245, the North Pole would be the first to 
cool down for the deposition of oxides, so that deeper surface- 
rocks would form at first at this pole. This deposition being 
of a large mass of matter would disturb the centre of gravity 

* 'Fluids,' p. 391 ; also Brit. Assoc. Rep. 1884, p. 723. 


of the globe, bringing it to a more northern position; the 
effect of which would be that the surface-rocks of the South 
Pole having the dense metallic matrix matter left nearer the 
surface would represent altogether a denser mean gravitation 
surface. The waters of the ocean being free on the surface 
would flow towards the denser system of the South Pole 
in equation with the density of lower matrix matter to 
restore general equilibrium. At a later period, when both 
poles were frozen, the aqueous evaporation-surface would be 
greater around the South Pole, as it is at present ; and, as 
the cold is assumed to be sufficient during this period at the 
South Pole to cause the deposition of snow, it would be 
greater in proportion to the surrounding area of evaporation. 
So that the South Pole would gradually acquire and henceforth 
possess the greatest accumulation of ice. Under these con- 
ditions, as regards deposition of aqueous vapour, the phenomena 
of excess of deposition of matter would be exactly reversed 
from the earlier state when the North Pole received the 
greater deposition of nebular matter, as before discussed. 

278. The effects of the great accumulation of ice at the 
poles may possibly be treated most demonstrably by con- 
sideration of the present conditions, which represent the 
terminal extreme effects of the present system of ice- formation, 
so far as the uniform cooling of the earth has advanced. At 
the same time we must recognize that the long interval, between 
the earliest and latest period of ice-formation, must have 
effected a great many changes upon the earth's surface, as 
the resistance of the crust has constantly become greater 
against deflection of its surface, by cooling through general 
distribution of internal pressures, due to the local compression 
of ice at the poles. Some suggestions with regard to this 
point will be made further on. 

279. Present Conditions brought about by Elevation of Ice. 
With this general discussion of active conditions present, we 
may now proceed to discuss the theory proposed upon these 


premises of the effects of the action of elevated masses of ice 
upon certain portions of the earth's crust. In the first place 
there can be little doubt that there is at present a great 
accumulation of ice at the poles of the earth. In the southern 
ocean this forms a wall about the Antarctic Circle, according 
to Sir James Ross, of seldom less than 200 feet above the 
sea-level, where icebergs are constantly detached *. In certain 
districts these are evidently of much greater height, as we 
find large icebergs floating with the upper surface said to be 
as much as 600 feet above sea-level, indicating a submergence 
of probably five times this depth, or 3000 feet. As these ice- 
bergs are detached from the front of the coast, it is quite clear 
that the ice must flow down from the interior, as in the ordinary 
glaciers ; therefore there must be heavy deposition of snow 
accumulating at the back of them. Mr. W. Hopkins has cal- 
culated that ice will just move downwards at one degree of 
inclination. In taking the inclination of ice over some 
mountain-valleys in the Grrindelwald glacier in Switzerland, 
which is a very flat glacier for about a mile, I found that the 
mean for the lower parts of this glacier was not less than two 
degrees. The southern ice-cap includes an area approximately 
equal to the entire Antarctic Circle, that is of about 700,000 
square miles. Taking Mr. W. Hopkins's estimate, one degree 
of elevation (as pointed out by Dr. Crollf), makes the altitude 
of solid ice about 24 miles in thickness over the southern pole. 
Such a thickness, assuming the ice by compression to take 
nearly the solidity of surface-water, would represent a po- 
tential force upon the earth's crust of say 3950 tons per 
square foot about this pole, or taking an area of 10,000 square 
miles of surface around the South Pole, a pressure upon the 
crust in this region would be maintained of over 3900 tons 
per square foot. 

* ' Voyage to the Southern Seas,' Sir James C. Ross, vol. i. p. 219. 
f ' Climate and Time/ p. 375. 


280. We may consider further that it is scarcely possible 
to suppose the ice a floating mass wholly breaking away 
at the coasts or that it rests upon a level plane, the probability 
being that the surface is extremely mountainous or irregular 
inland near the terminal glaciers ; in this case we must allow 
for greater friction on the motive plane, consequently for 
greater depths of ice before slipping can occur. If 24 miles 
of ice be assumed to continue in a static condition at the 
South Pole, above the symmetrical earth considered as a 
spheroid of revolution, it appears to be highly improbable 
that the crust, supported upon a liquid matrix, could resist 
this excess of pressure without deflection; and even if it 
should do so, the accumulation of ice still remains a constant 
factor until the resistance is overcome. We must further 
consider that in Mr, W. Hopkins's experiment the free surface 
of the ice was only a short distance from the artificial inclina- 
tion measured *. It is possible that in the case of ice hundreds 
of miles inland, in every way supported by surrounding ice, 
grounded on a frictional or even possibly a surrounding 
mountainous plane and at a much lower degree of temperature 
than in these experiments, downward movements would not 
be possible at one degree of surface-inclination. Under such 
conditions ice might be permanently retained, if the earth 
were sufficiently rigid, possibly at two or three degrees of in- 
clination, as it is in our inland glaciers. With such a local 
pressure upon a yielding sphere, which we assume the earth to 
be, we can scarcely imagine the possibility of its resistance. 

Further, ice in cooling increases in density, and we can 
form no exact conception of its rigidity at 100 Centi- 
grade, as it probably exists at this pole. Forbes's observations 
showed that ice moved downwards in glaciers with velocity 
somewhat proportional to its temperature f- Accepting all 

* Phil. Ma<?. 1845, vol.xxvi. 
t Forbes, 'Norway and its Glaciers,' 1853, p. 234. 



the conditions as active upon a deflectible globe, to cause 
reactions upon its crust, then the elevation of land, earth- 
quakes, and volcanoes could be easily explained. Similar 
conditions to those defined for the South Pole would hold at 
the North Pole, although at the present time less effectively. 

281. Unfortunately the poles of the earth cannot be 
reached for exact evidence of the above-assumed conditions 
of accumulation of ice, but we happen to have in Green- 
land a similar state active in a less degree and on a relatively 
small scale. Here the inland ice being elevated above the 
snow-line, vapour-currents are constantly condensing to snow, 
which as constantly accumulates. So that the greater part 
of Greenland is at the present time a complete glacier- 
mountain of possibly 7000 to 8000 feet of interior elevation. 
In the southern part we have at the present time land- 
surface, and here the coast is now known to be sinking for 
the space of 600 miles *. Exact measurements have not 
been taken of the rate of sinking; but ancient buildings 
upon the rock-islands are said to be sinking beneath the 
ocean-surface, so that experience has taught the native 
Greenlander not to build his hut near the water's edge f 

282. It is very possible that the rate of sinking is nearly 
proportional to the increase of weight of snow annually 
piled up inland. Such pressure as may be produced in 
Greenland, situated as suggested over a viscous system of 
matter, will act hydrostatically and be felt elsewhere possibly 
by elevation in Iceland or Scandinavia ; but as the pressures 
will combine with the general system of the polar pressures 
in acting upon the heated magma beneath, where this par- 
ticular pressure is most reactive at present, that is where 
the crust is least resistant, it is impossible to discover except 
by observation. 

* < Nature/ Sept. 20th, 1883, vol. xxviii. p. 488. 
t Lyell's ' Principles of Geology,' vol. ii. p. 196. 


283. It is desirable perhaps to prevent misunderstanding 
to make a slight detour with regard to the sinking of 
Greenland, pointing out, what is quite evident, that all 
sinking of land cannot be attributed directly to loading such 
land with ice, as all sinking surfaces are not so loaded. All 
that can be asserted is that accumulations of ice, where suf- 
ficient, will cause such sinking. The sinking of a district 
may occur through tipping of an area of surface, possibly 
through local resistance to polar pressures, which act in the 
horizontal direction if this offers less resistance than the 
vertical. Such, for instance, as that of the Runn of Cutch, 
adjacent to the delta of the Indus, where in 1819 a consider- 
able area of land sank in one district and simultaneously 
arose in another. Such tipping may also be noticed in the 
neighbourhood of volcanic islands where there is contiguous 
local elevation. Possibly also in some cases it may be pro- 
duced by the loading of the interiors of continents with sand- 
drifts produced by prevailing dry winds. The whole of the 
instances of depression are small and local, and do not in any 
way compare with the numerous great and nearly constant 
elevations such as that of the western coast of South 
America of from thirty to three hundred feet, for a distance 
of 1180 miles along the coast and for an unknown distance 
inland *, or that of extensive land-areas in Scandinavia, 
which I assume are due to the reaction of accumulation of 
ice at the poles upon a deflectible viscous substratum of rocks, 
and consider to be a necessary condition, in the present 
order of things, to maintain land-areas against the constant 
degradation from atmospheric causes. 

284. Where Ice-pressures will be most Active. To return 
to the accumulation of ice at the poles as affecting by 
hydrostatic reaction the assumed semi-liquid interior, and 
thereby causing the elevation of land in other parts of the 

* Chas. Darwin, ' Geological Observation,' p. 209. 


globe, we may be assured in the first place that such ele- 
vation will be subject to two important conditions of the 
crust of the earth : 1. The crust may be nearly uniformly 
rigid ; 2. The crust may be fractured in parts, or possess 
lines of weakness. We will take the first condition. 

285. If the crust of the earth is nearly uniformly rigid 
throughout its exposed parts, it will be still evident that if 
there is an ice-cap of some miles in thickness covering both 
the poles of the earth, this ice-cap will materially add to 
the rigidity of the parts that it covers ; for we know it is the 
property of ice by r'egelation to heal any possible fracture or 
strain that may occur from any cause, and thereby constantly 
to present a very rigid mass, particularly if it is retained 
in a close area or supported by irregular or rigid surface 
matter, as before stated. In this respect it possesses a 
property of rigidity not shared by the tertiary matter of 
the earth, as this remains faulted after it has been once 
fractured by an internal strain. 

286. If we imagine the ice crust to be maintained at its 
surface at an elevation equal to one degree over the Ant- 
arctic Circle, that is 24 miles in thickness at the South 
Pole, and to be possibly of great thickness at the North Pole, 
we may then suppose that the land-surface covered by this 
coating will be highly indeflectible* Under these conditions 
the greatest effects of the opposing pressures of the poles 
would fall more nearly upon the tropical regions, where there 
is the greatest surface curvature, and each meridian would 
represent as it were a bent bow under the excess of polar 
pressure ; therefore in the tropics there would probably occur 
the greatest plutonic or volcanic elevation. Now as we 
know also that the tropical area, from rainfall, is the area 
of most rapid denudation, and consequently of more active 
thinning of the crust, we ought also to find evidence of 
this being the area of greatest volcanic action, and this upon 
the whole is fairly consistent with observation. The following 


diagram, fig. 23, will give details of the conditions proposed. 
Let N and S be the poles covered by an ice-cap, E and W 
the equator, TT', T" T"' the tropical regions. 

Fig. 23. 

We should then expect the surface of the earth or the out- 
flow of volcanic lavas from internal pressure to rise to the 
greatest height at T to T", T' to T", although this would in 
all cases partly depend upon the viscosity and other conditions 
of friction within the strata through which the internal 
pressures must pass from the region of polar pressures. 

287. Taking the above conditions, there are many matters 
which at once strike one as relative. Thus the mass of ice 
at the poles, placed originally by vapour forces out of equi- 
librium with the earth's symmetrical gravitation system as 
a spheroid of rotation, exerts a force upon the solid crust of 
the globe, which it overcomes in proportion as it is insuf- 
ficiently rigid for resistance. In this manner the resistance 
becomes distributed, so that it is not only from internal or 
hydrostatic pressures, but by direct horizontal pressure upon 
the crust that we may have certain facilities of upheaval and 
deflection of the surface rocks. 

288. Further, if we assume a line of expansion over the 
tropical area, which the elevation of lower or interior matter 
of the earth and constant denudation really indicates, this 
would produce also perpendicular strains in the rigid materials 
of the earth's crust, which would follow the meridians, par- 



ticularly in the lines of original polar extension of the 
lighter elements of matter ( 250). Thus we should have 
brought about the conditions of lines of weakness where the 
pressures at the poles would not act in a direction to close 
them by any form of crumpling action, as they might be 
assumed to do in the latitudinal lines. This principle may be 
shown experimentally by the fracture of an india-rubber ball 
or a bladder filled with water or air by opposite pressures 
towards the centre through one diameter, as shown in fig. 24, 
wherein a fracture is produced from n to s. 

Fig. 24. 

From these causes it is probable that although earthquake 
regions exist around the globe as an effect of internal pressure, 
we have the greatest volcanic effects in the tropics, other 
conditions being equal, which may occur in lines at nearly 
right angles to the Equator, and continue poleward over land- 
areas as in the South American range. 

289. The above cause of fracture, and the lines of weakness 
pointed out, may generally rule the positions of volcanic 
elevations. But another cause of weakness or faulting 
may be suggested, which is particularly relative to the 
distribution of ice that is, that in the immediate vicinity of 
great accumulations of ice there will be great compression 
locally upon the immediate surface strata, and this reacting 
with less friction than at a distant part, may cause faulting 
near the area of pressure to a considerable depth, as at points 


a, a', a", a'", fig. 24. This may possibly account for volcanoes 
in the Antarctic and near Arctic regions of Erebus, Terror, 
and Skaptar-Jokull. 

290. Presence of Water or Steam common to Volcanic 
Eruptions. If ice-pressure were present at the poles, 
bearing down the surface rocks upon the lower viscous 
matrix and causing the displacement of the lower viscous 
stratum of tertiary rocks, the originally cooler surface rocks 
would necessarily come finally into contact with the central 
highly heated metallic core. We may imagine that when 
this occurred the surface rocks as they were pressed down- 
wards by superimposed weight of ice would be melted at 
their lower surfaces, and become themselves heated viscous 
rocks. These would be again displaced laterally into the 
general viscous mass by the superimposed pressure, until 
a thin stratum only of cool mineral matter would remain as 
a permanent non-conductor between the ice and the heated 
liquid metallic core beneath. The metallic core would also 
in time be chilled to a rigid coating by the constant arrival 
of cooler tertiary matter abstracting its heat. The weight of 
ice at the South Pole would constantly increase, particularly 
in antarctic winter, and this must continually react upon the 
surrounding viscous matrix beneath, and the surface of the 
slightly chilled metallic core, just as the former surface rocks 
had previously acted upon it, as there are no possible con- 
vection heat currents in the ice. In this case the former 
rock-pressure upon the metallic matrix would be replaced by 
a water-pressure acting through the thin stratum of rocks, 
which must necessarily continue to act as a non-conductor 
between the metallic matrix and the ice. Now as we assume 
that water as it was liquefied from the ice at the surface 
of the lower slowly conducting rock would receive more 
pressure above than resistance laterally, it would under- 
flow outwards from the sub-polar area and permit more ice 
to come in contact with the surface, which would be sufficiently 



heated for its liquefaction, so that the process of liquefaction 
and underflow would be continuous for the dispensation of 
polar pressures. Further, as the water would underflow 
from near the surface of the metallic core in an approximately 
horizontal direction, and at the same time be released from 
a part of the superimposed pressure in passing beyond the 
polar area, its tendency would be to flow upwards through 
liquid rocks nearer to the surface, which would be of less 
thickness further from the poles. Therefore, convection 
currents would be impossible backward to the earlier position 
of the melted rocks. In this manner the underflowing water 
might become highly heated by the lower semi-liquid rock 
through which it flowed. This process may be shown by a 
diagram, fig. 25. 

291. Let A be the highly heated metallic core of the globe, 
the mineral coating of the lower part of which is shown 
at x x, x' x" ; Na, Sa' the ice-caps. Then by the pressure 
upon the lower surfaces at x and #, the lower liquid heated 
matter unable to resist the pressures would be driven to 
equilibrium of symmetry with the earth's spheroidal form by 
underflowing in the liquid plane x x toward x' x'. Further, 
if the ice which formed the caps Na. Sa r by continuity of 
pressure penetrated to the surface of a non-conducting 



stratum at x x, where matter was at a white heat within a short 
distance from the polar region, it would dissolve and carry 
with it mineral matter from the lower surface rocks, which 
would be extruded at the first position upon the globe 
where it could overcome the resistance of surface rocks to 
establish static equilibrium. 

Fig. 26. 

292. Fig. 26 shows details diagrammatically of the manner 
in which water dissolving mineral matter could underflow the 
central surface-rocks of the globe. I represents the southern 
ice-cap. N the dense metallic nucleus-matter, VR the lower 
heated viscous rocks resting upon the nucleus-matter. CR 
the present chilled viscous matter forming the surface rocks. 
C the chilled surface of the lower viscous matter which serves 
as a non-conductor between the heated nucleus-matter and 
the ice. L the position where the ice-pressure becomes 
sufficient to force the lower water beneath it into the viscous 
matter. This occurs at a lateral position where there is the 
least resistance to the hydrostatic pressure. The water is 
forced to form a channel through the viscous rocks into 
which the remaining water below the ice-cap flows. As the 
water flows from the polar sub-basin it becomes heated and 


expands possibly to double its former volume and dissolves 
mineral matter from the channel it has formed. The aqueous- 
mineral matter being of less specific gravity than the viscous 
rocks, flows towards the surface of the globe, where it may be 
projected as volcanic matter, or by its hydraulic pressure 
float up the surface rocks. 

The action of release of polar pressures will be generally 
paroxysmal, as the channel once opened by the pressure will 
continue flowing through backward-pressure until the liquefied 
ice beneath the ice-cap is exhausted. The channel will then 
close until the hydraulic pressure again overcomes the resist- 
tance and causes it again to break through the chilled surface 
of the lateral viscous rocks. 

A channel originally driven through the viscous rocks 
would find ventage near the point of polar pressure, but the 
water at the same time would chill the surface-rocks from 
which it derived its heat. Under this condition the ventage 
would become consecutively lower owing to the surface resist- 
ance becoming greater, until, as at the present time, the mass 
of chilled matters shown in our diagram at CB prevent the 
outflow or projection of the aqueous-mineral matter until it 
reaches a great distance from the pole. 

The water forced to form a channel in the lower viscous 
matter, in becoming heated after a certain length of flow, 
would react upon the inflowing current through its expan- 
sion as a resistance. In this manner its injection would 
become intermittent. I arranged several experiments to show 
this and found one that would do so very simply. Making the 
stem of an ordinary clay tobacco-pipe red hot, and plunging 
it into water, the water enters the tube of the pipe inter- 
mittently, and is expanded and projected through the pipe 
into the air in separate spirts. 

The points of departure for the underflow channels, pro- 
jected by sub-glacial pressures from the Antarctic pole, must 
occur at the points of greatest extension of land where the 


subterranean heat can be best conserved by the more con- 
ducting nature of the surface rocks. Two such channels 
when they are open leave the Antarctic circle at about 60 
W. long, for the South-American range, and about 140 E. 
long, for the Sunda Isles. Where the land pressure is not 
great near the polar pressure this may continue to be a point 
of least resistance where a fault is once opened, as in Mounts 
Erebus and Terror in the great Antarctic bay, 170 E. long., 
and cause eruption at this point. 

In this manner a volcano would extrude viscous mineral 
matter only at its early stages, the backward pressure being 
upon liquid rock only ; but assuming that the volcanic crater 
continued to represent the point of least resistance to the under- 
flow of the system, the continuity of the volcanic outflow 
would open out a channel for the following extrusion of water 
from the pole, which would carry with it dissolved mineral 
matter, or what becomes, when reduced to atmospheric 
pressure, volcanic dust and steam. This might finally appear 
at such distant points as 6, b 1 , b". To find evidence of the 
above stated hypothesis, which I had previously introduced 
in a paper read at the Geologists' Association, March 1883 *, 
I examined very carefully under the microscope the volcanic 
dust that was blown from the Krakatoa eruption in August 
1883. I found it to consist principally of very thin spheroidal 
surface plates of volcanic glass (bubble plates as I termed them) 
of only about 73^0^ inch in thickness, resembling on a small 
scale broken watch-glasses. The vesicles, of which the bubble 
plates were the remains, were evidently originally filled with 
steam from the boiling mineral matter beneath, which expanded 
by release of atmospheric pressure, until they burst when cooled 
down high in the atmosphere. In this manner they added 
the effect of expansion to the polar pressure which propelled 
them through the neck of the crater, and therefore projected 

* < Nature,' 1883, vol. xxvii. p. 523. 


them to great heights in the atmosphere. The dust I examined 
was swept up from the deck of the bark ' Arabella/ sailing 
in the Pacific, at 1000 miles east of Krakatoa *. 

It is not necessary to discuss the manner in which 
heated water at high pressure dissolves rocks in detail. 
Water is proved experimentally at a temperature of 412 C., 
and at a pressure of 100 atmospheres, to occupy about four 
times its original volume, in which state it dissolves glass 
rapidly |- At a higher temperature it dissolves siliceous 
rocks, so that this could very well form an underflowing 
current of siliceous matter to the volcano, upon the principles 

293. In all volcanic systems, irrespectively of internal 
expansion of projected matter, there must be a tendency, upon 
the principles discussed, for the volcanic outflow to rise to 
hydrostatic equilibrium with the polar pressure. Thus such 
constant open volcanoes as Kilauea may represent safety- 
valves of this pressure ; but as this mountain is not so high 
as some other volcanoes which have extruded lava recently, 
it is quite clear that upon my theory the friction of the under- 
current of viscous mineral matter must withhold a part of the 
hydrostatic pressure caused by elevation of ice at the South 
Pole. This is also evident in the differences of altitude of 
Mauna Lao and Kilauea, which are near together. 

294. We may conclude that if a volcanic vent rises nearly 
to equilibrium with the distant pressure system, the exposed 
mass will cool quickly in the atmosphere, and the former 
point of least resistance may become the point of greatest 
resistance, and the volcano become permanently extinct, or 
it may overcome surface resistance nearer the base of the 
mountain than the original crater. If the volcano does not 
rise to the point of equilibrium it may become intermittently 

* Quart. Journ. R. Meteor. Soc. vol. x., July 1884. 

t Cagniard de Latour, Ann. de Chimie, ser. 2, xxi. & xxii. 


active with others in the same state, each eruption evidently 
clogging or preventing by the weight of ejected matter in 
the crater future eruptions for a time in the locality, and 
making the eruptions thereafter more paroxysmal. If we 
omit the friction of the system from consideration, then the 
pressure of the highest vertical column of the chimney of a 
volcano may be taken to represent the hydrostatic pressure 
upon the liquid matrix beneath, and from this the height of 
ice at the South Pole might be estimated. For this, Chim- 
borazo might be taken. But the evidence just quoted of 
Mauna Lao and Kilauea shows that friction may form a large 
factor of resistance to the distribution of polar pressures, so 
that equilibrium by ventage cannot be estimated. 

295. It is impossible within the limits intended for this 
work to discuss popular theories with which the above may 
at some time come into competition, but it may be well just 
to mention the two most popular. 

Firstly, that volcanic and plutonic phenomena are due 
to the shrinkage of the earth from cooling, which theory 
appears to persist in our text-books. This is fully worked 
out by the late Mr. Mallet in over one hundred pages of the 
Phil. Trans., 1874-5. But the whole matter is built upon 
inexact data, and there is such great error in the calculation 
that this must in time suppress this weak theory*. Observa- 
tions of stratified rocks give evidences almost universally of 
vertical separation by open cracks, which indicate quite the 
reverse of a horizontal surface pressure, such as would be 
the certain result of an internal contraction from loss of 
heat, demanded by Mallet's theory to cause the elevation by 
crumpling up of the surface rocks. The universal open cracks 
are, on the other hand, the natural result of the effect of 
elevation by plutonic forces as herein proposed. 

Secondly, the theory that volcanic phenomena are pro- 

* Appendix B. 


duced by the expansive force of steam, originally proposed 
by Spallanzani in 1788, but best known by its development 
by Scrope *. This theory could not have been proposed 
with a better knowledge of physics. The experiments of 
Cagniard de Latour before mentioned, in which water was 
made red hot at a temperature of 412 C. with an expansion 
of only four times its volume in a hard glass tube without 
bursting, demonstrate that water could not act effectively 
in the projection of rocks as in volcanic phenomena under 
a very moderate pressure of superimposed rocks, say one 
hundred feet of liquid basalt, where the water might exist of 
a white heat and of not over double its ordinary volume. Of 
course, if the rocks were dissolved in white-hot water, as they 
would be at this temperature, the water would expand into 
steam when the pressure was released by its coming to the 
earth's surface ; but this is very different from the assumption 
that steam at such pressure is able to cause the elevation 
of thousands of feet of solid rock. 

296. There is a further condition frequently offered to 
support the steam theory in the assumption, in opposition to 
hydrostatic laws or the observed conditions of heated rock, 
that water may percolate rock at a low level from the ocean 
and be projected at a high level where the superimposed 
pressure of surface rock is greater, which is evidently 
impossible. Neither are deep-seated rocks, as we find them 
in deep mines, porous enough to admit of such percolation 
even if it were sufficient to account for actual phenomena. 

297. The conditions stated in this chapter relate to deposi- 
tion of ice since the miocene period. There were possibly 
earlier depositions of ice, the conditions of which will be 
discussed in the next chapter. But I anticipate that all the 
greatest effects of polar compression by ice, and therefore of 
volcanic eruption and plutonic action, followed the miocene 

* Scrope on Volcanoes, 1825, p. 17. 


period, which, I think, from the greater variation of animal life 
and its higher development, was a longer period than any 
geological epoch that preceded it. This I will discuss later. 
Geologists are generally of opinion that volcanic action has 
been constant. I am not certain that the evidence is clear on 
this point. If a volcanic vent is found penetrating the 
silurian surface rocks, there is no reason why it should not be, 
in some cases open to observation, a tertiary one, as volcanic 
forces penetrate all surface rocks. Neither is it necessary 
that igneous matter should reach the surface, as at a certain 
pressure it may float up the surface rocks to form mountains 
without any extrusion ; and if these are afterwards degraded, 
the volcanic vent will appear as though it issued at the 
stratum level at which its intrusion occurred. At the same 
time there may have been exceptional conditions of ice- 
formation at early periods, which will now be considered. 

[ 210 ] 



298. Condition of the Sun during Earth-formation. 
Taking the earth system to have been abandoned by the sun 
as a nebulous zone according to the theory of Laplace, the 
earliest condition of the sun in relation to this zone would be 
that of a luminous globe of nearly the diameter of the earth's 
orbit. As the earth-zone commenced to condense and form a 
planet, the sun's volume would at the same time be also con- 
densing and leaving this zone more free by distance. If 
there was any inequality of density in the zone-ring, the 
most probable condition, there would then be a separation at 
the most attenuated part of this zone, and a further con- 
densation in another part by the continuity of the system of 
attractions of the denser matter. So that it is probable that 
the earth condensed into a nebulous globe by concentration 
before any solid matter was formed at its centre. 

After the detachment of a nebular planet-zone, this zone 
having much greater surface area relatively to volume in 
comparison with the voluminous sun, and being open to free 
radiation into space in all parts not directly facing the sun, 
would condense much more quickly than an equal volume 
of the sun's nebula. 

299. In considering the condensation of the nebulous zone 


as a special exterior formation, we may conclude that no 
condensation could occur unless the radiation of its heat into 
space exceeded that derived from the condensing sun, although 
at the same time the nebulous zone must have continuously 
absorbed the sun's heat falling upon it. Under these con- 
ditions the radiation of heat into space from the zone must 
have been of its initial amount plus that constantly received 
from the solar centre. If we imagine, what is most pro- 
bable, that at the early time of condensation by cooling a 
clouding would be produced in this zone through excess of 
radiation, then the sun's rays absorbed into the zone would 
be diffused within it, so that the radiation would take place 
from the zone in all directions, but more particularly in the 
exterior parts and those perpendicular to the plane of orbit 
where it was most open to free space. 

300. Under the above-stated conditions we may consider 
the effects of the formation of a nebulous planet-forming zone 
upon the amount of radiation of heat and light that the sun 
would be able to disperse beyond this impediment to an 
exterior planet, assumed to be fully formed at the time. 
Then, assuming the earth already formed, and the sun 
contracting in volume before the period of the formation of 
an inner planet, as Venus, a nebular band, or what we may 
term the Venus-zone, would appear across the sun's disc, 
which would continue to obstruct a large part of his rays, and 
this would last until the complete formation of Venus as a 
planet. The like would again occur before the complete 
formation of Mercury. We have apparently, upon a large 
scale, nebulae in this condition *. 

After a nebular planet-zone was detached from the sun 
its future condensation would depend upon contingent 

In fig. 27 the possible appearance of the large nebulous 

* 2244 Gen. Cat., Rosso Nebula, p. 90. 


sun is represented partially obscured by the Venus-nebular- 
zone at a certain stage of its condensation. 

Kg. 27. 

301. If Venus or Mercury, from inequality of distribution 
of nebular matter after the time when their zone-rings were left 
by the sun or from any following disturbing cause, as the 
intrusion of a comet, condensed at first into a nebular globe, 
as just proposed for the earth ; then this globe, after its 
formation, would in time concentrate to an intensely-heated 
nucleus at its centre. This nebular system would obstruct 
the sun's light through cloudiness at its early period of 
formation ; but afterwards for another following period, 
when it became incandescent, it would present the same 
light- and heat-giving radiation as that of the nebulous sun 
to the surface of an exterior planet, or greater in proportion 
to its visible surface and its state. This new-formed planet, 
as before stated, would therefore be obstructive to light and 
heat from the sun if it was nebulous, or auxiliary to it if it was 
incandescent. In either case this excess or defect of light 
and heat would occur in periods of the newly-formed planet's 
synodic revolution in relation to an exterior planet, producing 
in either case intermittent periods of intensity of radiation of 
light and heat upon the exterior planet. 

302. After the periods of the entire condensation of a 
newly-formed planet to non- obstructive or non-auxiliary heat- 


giving proportions when it could not affect the amount of 
heat and light dispersed to an outer planet from the sun, the 
sun would return to its normal luminosity according to its 
age or state, and become again the only source of heat-energy 
to the planet. 

303. It is not necessary to assume that the sun, when it was 
an immense nebular globe, gave out in its entirety more heat 
and light than it does at the present time ; it very probably 
may have given out much less at certain periods. The super- 
ficial condensation of nebulous matter upon or about its outer 
surface would constantly cloud the effects of its intense 
internal heat, so that when freed from the clouding influence 
of the formation of a planet-zone, it would present to the earth 
only the appearance of a magnified image of one of our stellar 
nebulae*, or of a large bright cloud. 

When the sun's nebular diameter was as great, for instance, 
as the diameter of the orbit of Venus, if its surface heat had 
been of the same intensity as at present, condensation of 
nebulous matter to form the earth would have been impossible. 

304. Distinct Solar Heating Periods. Under the conditions 
just stated it will be seen that the sun's heating effects as 
regards an exterior planet, as, for instance, the earth, would 
pass through certain phases or periods from the time of the 
formation of one interior planet to that of another more interior. 
Thus with regard to the earth we should have one long 
period during which the sun would be slowly decreasing in 
volume, appearing during this time as a large bright nebulous 
globe or stellar nebula. This condition of the sun would 
continue, during its condensation, until the time of the dis- 
turbing conditions of the clouding effects upon its disc of the 
commencement of the formation of the Venus-zone of nebula, 
which would, after a time, appear to entirely encircle the 

* See 4883 Gen. Cat. Ros. Obs. p. 170, also M81 and tf 1205 Ursaj 


nebulous sun as a dark band. The local condensation 
within the nebular band would ultimately form Venus into 
a large globular nebulous planet, the clouding-effects of 
which when moving over the sun's disc would diminish the 
sun's heat in transit and make it therefore intermittent in 
intensity with regard to the earth, as just stated, in periods of 
584 days. This passing of the nebulous planet over the large 
nebulous sun at the time of inferior conjunction might 
possibly at first nearly obscure his light and heat, at other 
times, exceeding nine tenths of the period of revolution, the 
nebular sun would appear bright and open. 

305. It will be seen on examining the conditions just pro- 
posed, which must be incidental to inferior planet-formation 
as regards the earth, that there were nine somewhat distinct 
periods of minus and plus solar radiation, therefore of greater 
or less heat and light radiation, affecting the formation or 
depositions of matter upon the earth's surface. These may 
conveniently be defined, to show the effective state of the sun 
as a radiating body under the entirely nebular conditions 
proposed during the formation of Venus and Mercury as 
far as the}^ would affect the earth. 

1. Period of open radiation of bright nebulous light and 
heat lasting from the separation of the earth-zone from the 
sun until the commencement of the condensation of the 
Venus-zone, during part of which period the earth was a 
nebulous globe. 

2. Period of nebulous obscurity of the sun caused by an 
absorption band across the sun's disc ; period of dull nebulous 
light and heat lasting from the early part of the condensa- 
tion of the Venus-zone until its formation as a nebulous 

3. Intermittent light and dull periods of about 584 days, 
the bright periods much exceeding the dull periods and in- 
creasing in brightness from the time of the condensation of 
Venus to a globular nebulous planet until it became non- 


obstructive to solar radiation. Period of great disturbance 
of local conditions of deposition upon the earth. 

4. Period of intermittent excess of bright nebulous light and 
heat caused by the presence of Venus as an incandescent 
body, particularly near the period of transit, lasting from the 
time of condensation of Venus to an incandescent liquid or 
solid planet until it ceased to be in any degree auxiliary to 
the sun's heat. 

5. Second period of bright open light of the sun, lasting till 
the commencement of the formation of the Mercury-zone. 

6. Period of nebulous obscurity by a dense band across the 
sun's disc, lasting from the early part of the condensation of 
the Mercury-zone until the complete formation of Mercury 
as a nebulous planet. 

7. Second period of intermittent light and dull periods of 
about 116 days, the bright periods much exceeding the dull 
periods, the light increasing with time, lasting from the time 
of condensation of Mercury to a nebulous globular planet 
until it became non- obstructive to solar radiation. Period of 
local disturbance of systematic stratification of disintegrated 
matter upon the earth. 

8. Second period of auxiliary light and heat, when Mercury 
became an incandescent globe, lasting until it ceased to add 
to the sun's light and heat. The whole period much less 
active than the corresponding period of Venus condensation. 

9. Third open period of bright light, lasting from the com- 
plete formation of Mercury as a non-auxiliary light-giving 
planet until the present period of intense solar radiation and 
for all future time of effective solar energy. 

306. Influence of Inclination of the Orbits of Inferior Planets 
and Eccentricity. It will be seen with regard to the Venus- 
zone and the second period of dull nebulous light, that owing 
to the inclination of the plane of the orbit of Venus, 3 23', 
unless the nebular zone had a sectional diameter of about 
10 million miles transverse to this plane, it would not 


continuously cover the centre of the sun's disc, so that it 
would appear to shift about from north to south in synodic 
periods, the entire variation of which would take about 243 
years. This would occur until its condensation into a nebular 

The same conditions as described above would occur in the 
sixth period with the Mercury-zone; but in this case we have 
an inclination of orbit of 7 with great eccentricity, so that 
the centre of the sun's disc would be covered by a nebular 
band of 5 millions of miles, and if it were of this diameter 
transverse to the plane of orbit, its effects upon the diminished 
solar disc would be nearly the same as that of Venus, but in 
shorter periods of 7, 13, or 46 years. Under these conditions 
the dull periods proposed would be in degree intermittent and 
produce great changes in the atmospheric conditions of the 

307. At the formation of a nebulous globe in the early 
part of the sixth and seventh intermittent periods, it is pos- 
sible that each of the planets Venus and Mercury, when a 
nebula, subtended as great an angle at the earth as the nebular 
sun of these separate periods. In this manner the sun would 
be obscured at every inferior conjunction and have its disc 
partly obscured for nearly one tenth of the particular planet's 
synodic period. This intermittent condition would probably 
produce greater effects upon the earth's atmosphere, and 
therefore upon deposition of rocks, than the dull periods, 
second and sixth, previously considered. 

308. It is scarcely possible to realize the enormous effects 
that would be produced by the obscuration of the sun for a few 
days only, but very possibly this was only partial. If actual, 
nearly the whole of the waters of the vapour-laden atmosphere 
would be precipitated. Terrestrial organic life would be 
destroyed by the sudden cold. Inland lakes and rivers, and 
part of the ocean, would be frozen, so that an entire organic 
change in animal life might follow, this being possibly 


afterwards evolved from the preservation of former marine 
and aquatic species that could retain life only by remaining 
deep in the ocean. So that, so far as this proposition goes, 
evolved species may not be found to be by any means 
uniformly progressive from the highest types throughout 
geological time, that is, even for the known progressive types. 
309. In the above scheme we have considered the effect of 
the condensations of Venus and Mercury, which were possibly 
largely induced by the reduction to the critical temperature 
of the nebular matter which formed the central solar system, 
95. It is probable that this formed only a part of the 
changes in the amount of heat radiated from the central solar 
system to the earth. The conditions of critical temperature 
of the sun have been discussed. The condensation which was 
capable of detaching a planet-zone would at the same time, as 
before proposed, obscure a large part of the heat of the central 
system by metallic cloud. Further, the extensive critical 
condensations which would form a planet-zone would be ex- 
ceptional. There were most probably many minor conden- 
sations at the critical temperature of certain elements which 
produced no detachments from the central solar system, but 
only partially obscured his light and heat. This would cer- 
tainly occur within the long period following the condensation 
of Mercury. Possibly such a condensation may have pro- 
duced our glacial period. Even at the present time the sun's 
radiation may vary greatly in long periods depending upon 
the sun's surface density, that is, the amount of condensation 
at critical temperature tending to cloud the chromosphere. 
Under any condition the interval period of open solar radiation 
caused by the simple contraction of the sun's volume when no 
effects of planet-formation were present would, in all proba- 
bility, exist for a much longer period than that of any planet- 
formation. This will be reconsidered. It is only important 
now to recognize the general effect of the contraction of the 
sun upon the earth in its more extended action. 


310. General Effects of the large Nebulous Sun upon the 
Earth's Meteorological Condition. Perhaps the most distinctly 
important condition that would be due to a large nebular sun, 
free from the disturbing effects of planetary formation, would 
be that the whole of the earth might be diurnally lighted and 
heated by the sun, and that this diffusion of light and heat 
would establish very calm conditions in the atmosphere. 
The tropics would not be excessively heated and the polar 
regions would receive at all times direct sun's rays, so that, 
assuming the body of the earth had cooled below a temperature 
to affect the evaporation of the oceanic surface, there would 
not be the excess of evaporation and expansion of air and 
vapour over the tropics as at present, due to solar radiation, or 
the great excess of condensation in polar areas, to produce 
the violent atmospheric currents we have at present in 
storms and cyclones. 

311. At the period when the sun had condensed to the 
orbit of Venus, the diameter of his disc would exceed 67, so 
that it would present a luminous surface of this angle, and 
the poles, even in mid-winter, would have a segment of the 
sun shining daily through an arc of 10 versed sine. At the 
period when the sun had decreased to the dimensions of the 
orbit of Mercury it would still present an angle of 44. There- 
fore, even in mid-winter, there would be at least a daily twilight. 
Altogether the general conditions of the globe would be very 
equable as compared with the present, although, of course, the 
polar area would still be colder, as being more open to free 
radiation. If, on the other hand, we consider in contrast the 
conditions that would be active during the shorter intermittent 
periods 2, 3, 4, 6, 7, and 8, 305, the atmosphere would at 
these times be subject to constant disturbance by high winds 
and heavy rains in intermittent hot and cold periods, which 
would cause rapid decomposition of rocks, and generally 
produce intermittent thin planes of varied stratification. 

312. As regards depositions of matter upon the earth from 


degradation of rocks, these must depend always in amount upon 
the atmospheric conditions present. We might have a long 
period during which depositions went on very slowly. A 
period of light rains and nearly constant sunshine, in which 
very little matter would be detached from the rocks or 
brought down to the sea-level, during which time, the con- 
ditions of life being constant, there would be very little reason 
for any change of form, and the strata of universal matter 
deposited, although uniform, would be very incommensurate 
with the time. On the other hand, we might have periods of 
heavy rainfall and frost in high lands, during which time de- 
gradation and deposition would be very rapid and the strata 
formed be very deep for the limited time of formation. With 
great change of atmospheric conditions there would be great 
struggles for existence, in which favourable varieties of life- 
forms only would survive. 

So that, on the whole, no depth of geological stratification 
nor even changes of animal life will indicate, with any exact- 
ness, the extent of time of a geological period unless we possess 
other data for consideration. 

[ 220 ] 



313. Time of Condensation of our Solar System. This may 
possibly be found with some degree of approximation by 
estimating the rate of contraction of the original nebula upon 
thermodynamic principles. For this calculation we may 
take the dimensions of the nebula at any period of its con- 
traction when it occupied a space within the extent of our 
solar-planetary system until the present time, when it has 
contracted to the small volume of our sun. In this propo- 
sition we again accept the probability that the nebula was 
originally one of the symmetrical planetary nebulae of which 
we possess many instances open to astronomical observation. 
A probable form being that of one of the nebulae shown, 
Plate II. a, 6, c, or d. Such a nebula under the conditions 
we have assumed throughout this work, must have condensed 
constantly by radiation of heat from its surface, and at the same 
time have formed our sun by concentration of matter about 
the centre. We may form an estimate of the extent of this 
nebula, at least at a certain period, by the extent of the 
extreme planetary orbits of our system ; and if we knew the 
rate of radiation at the surface and of concentration of matter 
at the centre or sun as a heat focus, we could then estimate 
the time the system may have taken to arrive at its present 


314. If we omit the consideration of the reaction of con- 
densation of heat in the central nebula or sun, and regard 
the radiation as a superficial function only, this will dispense 
the mean energy of the system in proportion to the tem- 
perature and extent of surface. If we assume that the 
heat was equally distributed throughout the volume of the 
nebula, the surface-heat would be nearly maintained by the 
superficial contraction and by heat exchanges with the 
interior of the system. Taken in this manner, a voluminous 
nebula would condense to less depth superficially in a given 
time than one of smaller volume. 

315. Observations of planetary nebulae, as, for instance, 
$ I. 205 or M 81 Ursse Majoris, Plate II. b and d, show a 
central condensation of large volume surrounded by a more 
attenuated medium. Under this condition, which we have 
assumed throughout this treatise may represent an early 
stage of our solar system, we have necessarily two distinct 
systems of condensation that about the solar centre which 
condenses its matter from the more attenuated surrounding 
medium, and that at the surface of the medium itself, which is 
contracting in space. The final condition of such contractions 
we assume will form a sun or star, that is, a unit incandescent 
mass. We may assume that our sun has arrived at nearly 
its final state. The only representative of the former nebular 
state being found in the chromosphere, and the largely diffused 
surrounding matter of which we have evidence in the corona 
and the zodiacal light. 

316. The amount of condensation of the surrounding 
luminous matter or pneuma into the central solar nebula 
cannot be defined ; but as the condensation must have pro- 
duced heat at the surface of the interior nebula, which again 
reacted by radiation, it is not probable that the interior con- 
traction was greater at any time than it is at present. For in 
the first case, the solar contraction would be blanketed by 
the surrounding pneuma, and in the present state its heat is 


freely dispensed into open space. Therefore we may con- 
veniently consider the present contraction of the sun as a 
periodical constant without great risk of error, for which we 
may obtain some data from thermodynamic laws. On the 
other hand, the contraction of the attenuated medium or 
pneuma, which formed the limiting volume, as this has 
practically disappeared from the solar system, must have 
proceeded at a higher rate. For this calculation we may 
indulge in certain assumptions which may give approximate 

317. According to the original calculation of Helmholtz the 
sun's radius is diminishing at the present time by 1/10,000 part 
in 2000 years =1/20,000,000, or 126*32 feet annually. This 
estimate has since been slightly increased. For an even 
number we will take 130 feet, which for convenience may be 
considered as a constant of time-condensation of the solar or 
nucleus nebula from the early period which we are now 

318. As regards the limiting superficial contraction of the 
pneuma, we will suppose that it condensed in some degree 
in proportion to its tenuity or distance from the solar centre. 
At a distant position it may have been rapid ; on the sun at 
present it may have nearly ceased. We will take it that at the 
period when the solar nebula extended to the orbit of Neptune 
the mean annual decrease of its radius was, as an extreme 
condition, possibly equal to two miles. Upon these data 
time elements may be calculated. 

319. For the time of condensation of the solar nebula to 
the present condition of our sun we will call the radius or 
mean distance of the orbit of Neptune, in feet, r ; the 
radial contraction of the solar nebula equal to two miles, or 
10,560 feet,/; and the constant of central solar contraction 
S = 130 feet. It is then clear that / diminishes, as it is 
assumed to do, at a uniform rate until it has vanished at the 
present time, its mean diminution was half this rate. There- 


fore the mean total diminution of the nebula for the limit of 
nebular condensation from the period will be *!+> This 


divided into r gives "7 =, time; which we find to be 2723 

millions of years the suggested time of condensation of 
the solar nebula from the dimensions of the orbit of Neptune 
to those of our present sun. 

320. Taking the same formula for the earth, calling its 
orbit-radius or distance from the sun r', or about -fa of r, we 

have 7 - =t f , which gives t' (earth-time) about 1008 millions 

of years, or, roughly, 1000 millions of years, for the time of 
the condensation of the solar nebula from the earth's orbit 
until the present time. 

321. The above calculation may be taken as quite the 
inferior limit of time ; no allowance has been made for the 
condensation at the solar focus which ultimately formed our 
present sun, which might increase in volume directly as the 
increase of gravitation from nearness of the more refractory 
matter to this centre, possibly in inverse proportion to the 
distances of all parts of the nebulous matter from the centre. 
The increasing compression of matter about the centre would 
more perfectly conserve the heat of the system, surrounded, 
as it would be, by a denser nebular atmosphere according 
to Lane's law, 11, p. 11. Therefore it is not probable 
that the decrease of volume from the effects of outer surface 
radiation of the nebula would progress at quite so high a 
rate as that just stated. 

We will now consider the conditions of time-variations of 
heat and light from the sun during its condensation, from the 
period when we assume it to have been a nebulous spheroid of 
the radius of the earth's orbit until the present time. 

322. Distribution of Time upon the Earth throughout the 


varying periods of the Condensation of the Sun and the Inferior 
Planets. This may be taken from the calculation just given 
of 1000 millions of years, making the divisions of time by 
fixing certain radii of the nebulous sun, to correspond with 
periods previously defined, for the probable clouding and 
auxiliary effects upon the sun's exterior radiation during 
planetary formation. In this proposition we may possibly 
divide the period of radiant energy into four classes con- 
sistent with our table. We may take solar energy to be 
open to be obstructed to be intermittently obstructed and 
open or to be increased by the energy of the condensed 
planet when this was in an incandescent state. The duration 
of these periods was probably proportional to the rate of 
condensation. I will take as hypothesis the formation of the 
nebular zone of Venus to have been obstructive to solar 
radiation during the sun's contraction for one fourth of the 
distance between the orbits of the earth and of Venus. That as 
soon as the sun's nebula was quite free from the Venus-zone, 
Venus would commence to form a nebulous globe. We will 
suppose that this nebulous globe remained intermittently 
obstructive to the sun's rays until the sun had contracted to 
one eighth the distance between the orbits of Venus and of 
Mercury. That Venus then gradually contracted and became 
by incandescence a bright body of possibly not more than 
six times its present radius. At this time it would be com- 
parable with the sun in density, and probably for a short 
period much brighter than the nebulous sun, even possibly 
as bright as our present sun. This state of Venus would 
slowly cool down, and possibly when the sun had contracted 
to one fourth the distance between the orbits of Venus and 
Mercury, Venus had ceased to become in any measurable 
degree a heat- and light-giving factor to the earth. After 
the condensation of Venus as a cool planet shrinking to 
nearly its present dimensions, the sun would remain open until 
the condensation to form Mercury. Mercury would then go 



through changes similar to those of Venus in proportion to its 
mass and its distance from the sun. 

323. Taking the above-stated condition that the contraction 
of the sun's nebular radius remained proportional to the time 
of condensation, we may construct a table for which we take 
1000 millions of years, previously given, as the entire interval 
between the earth's separation from the solar nebula until 
the present time. 


Conditions of the Nebulous Sun at its radius in Millions of 
Miles for nine Periods in Millions of Years. 


Radius of sun 
in million miles. 


Time in 
million years. 

1. Open 

93 to 73 



2. Dull 

73 to 67 



3. Open and Dull 

67 to 63 



63 to 60 



5 Open 

60 to 37 



6. Dull 

37 to 33 



7. Open to Dull 

33 to 31 



8. Auxiliary 

31 to 30 



9 Open 

30 to present 



This table may possibly continue nearly proportional if we 
consider the specific heat of the solar nebula less and its 
radiation greater, or vice versa factors that cannot be exactly 

324. Period of the Formation of the Nebulous Earth. If 
we take the period from the separation of the earth-zone, as 
proposed, 298, and assume this nebular zone was ten 
millions of miles in cross-section of the annulus, this would 
give upon condensation, if all its parts moved with equal 
angular velocity, an excess of rotation to the condensed globe 
produced therefrom over the revolution of the moon and 



the rotation of the earth at the present time ( 146) ; but 
as we presume that this zone would be partially condensed to 
discrete matter from exterior matter drifting in spiral paths 
thereto (222), which in falling upon the earth would tend to 
cause a negative rotation, the cross-section of the zone-ring 
proposed above may not be too great. This zone after 
detachment could not maintain its heat to the extent 
formerly suggested for the surface of the nebulous sun, as we 
have no large intense centralized source of internal heat 
comparable with that of the sun within the zone-ring, so that 
its heat would depend upon its condensation only, and, 
as this zone would possess much larger surface, relatively to 
volume, than the sun, it would be open more freely to radiation 
and contract at certainly quite double the rate of the sun. 
This would, as before stated, occur principally at its outer 
surface away from the sun, and at all outward angles trans- 
verse to the plane of orbit, as no central heat from the sun 
could fall upon these parts. In this manner the trans- 
verse radiation of heat from the zone would make the total 
contraction of the zone proportional to its sectional radius. 
Upon this proposition taking the sun's nebulous contraction 

at the position of the orbit of the earth as before proposed 

= ~ + S, which will be 482 feet in a year, or at the position 


of the outer surface of the ring about 500 feet in a year, 
denoting this by a, the number of miles in the ring-section 
by 6, the number of feet to a mile by c, and dividing by 2 for 
the contraction of opposite sides of the ring and again by 2 to 

make the contraction double of the sun's, we have - or 


=26,400,000 years, for the period of con- 

traction of the earth's nebulous zone upon itself to the position 
of the axis of the ring. 

325. In the above-stated proposition the contraction of the 


earth's zone-ring upon itself can only be considered theoreti- 
cally as the modus operandi. To give the moon its revolution 
period and the earth its rotation, we must imagine a separa- 
tion to have occurred at some part of the nebular zone, under 
which condition, simultaneously with the contraction of the 
section of the ring, it was being drawn together by gravita- 
tion acting in the linear direction of its circumference so as 
to form a nebular globe. This would, of course, materially 
complicate the conditions of the calculation of time by 
a differentiation difficult to follow ; but as the radiation- 
surface would certainly be constantly lessened during the time 
of globular condensation, we may prolong the time possible 
by nearly double the amount of the time suggested, say, to 
50 millions of years, for the condensation of our earth from 
a nebulous zone to a liquid globe at an intense white heat, 
surrounded as it must necessarily have been by an extensive 
nebulous atmosphere. This globe would possibly be formed 
principally of iron with the presence of other highly refractory 
metals, and become the permanent nucleus of much the greater 
part of the volume of our present earth, upon which the 
deposition of oxidized matter may have been superimposed 
upon principles already discussed. The superficial conditions 
are represented by the geological periods of the earth, to be 
considered in the following chapter. 


[ 228 ] 



326. Geological Periods. Leaving out of consideration 
any notice of the early paroxysmal school of geology, the 
eminent geologists, among whom Murchison, Sedgwick, and 
Miller may be particularly distinguished, who have made 
attentive study of a single group of ancient rocks, have 
come to the general conclusion that there have been special 
periods in the past which have been conducive to the formation 
of special kinds of rocks, which, with the fossils therein con- 
tained, are very distinguishable from other periods. That in 
and during the different periods for an immensity of time, 
wide areas of the globe were affected by like conditions from 
causes unknown. So that if we take, for instance, as the 
most striking example, the Silurian period of Murchison, this 
appears to be characterized by the presence of finely-deposited 
rocks in an entire broad band of unequal thickness surrounding 
the Northern Hemisphere, containing fossils of a similar 
succession of faunas, marked by particular zones of genera 
and species. It is probable also, judging from rock-texture 
and organic remains, that similar general conditions affected 
the land-areas of the Southern Hemisphere during the same 
lengthened period. 

A school of thought, founded by the genius of Hutton and 
developed by Lyell, wherein due recognition, not previously 
taken, is made of contemporary formations, has arrived at 


the conclusion that the present agents at work would, if 
active in the past, account for all former conditions of 
deposition. The theory hidden in the argument offered appears 
to be that of the improbability of other changes being active 
upon the earth than those due to successive astronomical 
and physical conditions, which are still active, and to the 
differences of position of land and oceanic areas upon the 

327. Quite outside all theoretical induction is the work of 
the practical or field geologist, who accumulates facts derived 
from observation irrespective of theory, and makes through 
observation only a broad distinction between the more 
ancient rocks, which were in a wide expanse special and 
uniform per stratum, and the modern rocks such as are at 
present forming, which, excepting possibly in the deeper 
part of the ocean, are locally only in very narrow limits 
similar to each other and vary in character in separate locali- 
ties in every degree. So that, taking the mean of geological 
opinion upon the early conditions, there appears -to be the 
extreme probability of the action of conditions in the cosmic 
system which have entirely passed away. These I propose 
now to consider as the effects of the astronomical changes 
already proposed in previous pages which have remained 
more or less evident in the early stratified rocks. 

328. It will not be found practicable in this .essay to con- 
sider stratification of rocks in detail. This has been well done 
by many scientific specialists in our advanced elementary 
treatises on geology, of whom it is only necessary to mention 
such names as Lyell, Dana, Geikie, Le Conte, and Dawson. 
It will therefore be necessary to consider in what follows 
the system of forces already proposed, which were probably 
active on a large scale as prime movers throughout the distant 
past under certain special conditions. There must also at the 
same time have been conditions which were constantly active 
affecting the deposition of mineral matter upon the earth. 


Of this may be mentioned the constant decrement of the 
sun's volume, the condensations of the solar- nebular, matter 
at critical temperatures, the astronomical constants of variation, 
of eccentricity of orbit, precession of the equinoxes, and change 
of obliquity of axis, and possibly also some changes due to the 
revolution of the magnetic pole which have not yet been 

329. By a general consensus of geological opinion the 
conclusion is arrived at that the long periods of deposition of 
surface-rocks which are open to observation can but be 
representations of detached units of geological time. This is 
shown most clearly in the great changes of animal remains 
between one stratum or set of strata and the next, in which 
the lost period of evolution often appears to be much 
greater than the long period made evident by the slow 
deposition of a formation or indeed sometimes of a single 
stratum. That this should be so may be inferred from 
the slow but constant action of meteorological forces alone 
in disintegrating rocks, which must always have been 
active upon the rocks protruded above the oceanic surface, 
and have produced continuous contiguous deposition of 
these rocks in another form at a lower level. This deposi- 
tion, which is general over the floor of the ocean, does 
not necessarily at any period become visible as surface-rock 
unless it is elevated by plutonic forces above the level of 
the oceanic surface. The evidences of the fossil and other 
remains in the rocks, which remain permanent as it were 
by the accident of being projected above oceanic level, show 
in many cases that our visible surface-rocks may have been 
elevated locally by plutonic forces and have been degraded 
by meteoric forces many times before they produced any of 
the present finely disintegrated strata. The materials of the 
rocks appear after disintegration to be sorted out as it were, 
so as to depart entirely in structure from their original 
character, and therefore they must lose their former history. 


The whole vestiges that remain of the long periods of the 
geological past represented by detached units, generally of slow 
deposition of mineral matter brought down by the action of rain, 
snow, wind, and tides, together with the large accumulation 
of remains of animal and vegetable life that have escaped 
after degradation, amount altogether to some 18 to 20 miles 
in thickness of deposition only, and these in separate detached 
units are all we possess from which to draw any inference 
whatever of the long period of past geological time. 

330- The above statement, as regards the extent of past 
geological time in its entirety, does not preclude our recog- 
nition, if we please to accept it, of the evidences of certain 
conditions that ruled for certain long periods for which we 
may attribute active causes. Thus, as an instance, during the 
great Silurian period before mentioned, we find locally even 
miles in thickness of finely and slowly-deposited rocks in 
even stratification in a system which appears to have extended 
to a greater or less depth over the larger part of the Northern 
Hemisphere. In this we find here and there ripple-marked 
surfaces left of the ancient quiescent oceans, and of rain- 
marks on the smooth sandy beaches of the period. We are, 
therefore, upon these premises bound to conclude that during 
this time we have a period of mild quiescent atmospheric 
conditions and of the minimum effect of tidal action, possibly 
with heavy rainfall, for which causes may be fairly suggested. 
This subject will now be considered generally by taking the 
separate early periods defined by modern geologists and 
endeavouring to correlate them with the periodic conditions 
which have been proposed, particularly with regard to the 
action of the effective radiation of the sun's heat and of his 
contemporary volume under circumstances which have been 
already discussed. 

331. Archcean Period. After the metallic nucleus of the 
earth was formed ( 235), we should have the possible approach 
of lighter nebular matter and its union with the surrounding 


oxygen and halogens present, entailing complicated chemical 
processes, the principal factors of which would be the con- 
densation and deposition of oxides and haloids upon the 
cooler polar areas of the earth, as before, proposed. The great 
atmospheric pressure would cause also the heavy rains of 
highly heated water in circumpolar areas as before stated. 
The hot rains under the great pressure would dissolve the 
siliceous, aluminous, and calcareous rocks that were already 
precipitated from the condensed nebula near the poles and 
those also that were crowding outwards in pressure-folds 
under the influence of gravitation whilst floating upon the 
central dense liquid metallic base-matter. By the hot rains 
the protruding rocks as they were constantly dissolved at 
their surfaces would be brought down into the hollows. 

332. Afterwards, as the atmosphere cooled, the semi-liquid 
rocks would be gradually deposited in a crystalline form in 
hollows and lake-basins. These depositions from aqueous 
solution probably produced the foliated crystalline rocks of 
the period most largely in gneiss, but partially also in mica 
and hornblende schists, chlorite slate, and crystalline lime- 

333. The land-areas formed of newly-deposited rocks at 
the close of the Archsean period would be left elevated by 
outward pressure of plutonic forces through polar pressures 
to great height at a distance around the poles. These rocks 
being constantly overflown with hot water left a system of 
permanent rocks, as before stated, locally thrown up at any 
point of minor resistance upon the first thin crust of the earth, 
or retained at the lower levels deposited from solution from 
its saturated mineral waters. The constant outflow of the 
lower heated viscid rocks moving into equilibrium to gravita- 
tional position, caused the upheavals from resistance to take 
place more particularly at the rounded borders of continental 
lands, from causes already pointed out. The general land- 
surface, therefore, was thereby covered with lake-basins formed 


by surrounding mountains most elevated at the polar sides. 
These basins as the temperature lowered held the deposits of 
mineral matter brought down by the hot rains, particularly 
from the poleward slopes, where rainfall would be heaviest. 

334. The entire system of what we now term the Archaean 
rocks is here assumed to be of the primitive rocks, which 
formed by themselves, and by after-degradation and ultimate 
deposition, the entire surface-rocks of the globe, as there 
could be no further deposition of mineral matter after the 
deposition of oxides and haloids from the nebulous atmosphere. 

This system of rocks, including the deepest or lowest stratum, 
which still probably retains a white heat, could not have been 
less than 100 miles in thickness. The chilled surface-rocks 
of the Archaean system, left measurable at the present time in 
Canada, the Outer Hebrides, Bohemia and Bavaria measure 
about 40,000 feet in thickness. 

335. Archaean Time in relation to the Conditions of Animal 
Life. As the lower or early-deposited basic rocks not exposed 
to surface-radiation would remain for a long period in a hot 
viscous state, they would continue to flow slowly to a position 
of gravitation-equilibrium, distorting and crowding up all 
the new cooler incipiently-formed strata. It is during this 
period, which would have extended until the commencement 
of the clouding effects produced by the condensation of the 
nebular zone of Venus, that I propose to place the complete 
formation of our recognizable Archaean rocks. This period 
possibly lasted altogether from that following the liquid 
condensation of the metallic core of our globe for about 
84 millions of years. This would complete nearly the entire 
first bright period, during which time the earth would 
remain in too highly heated a condition to maintain life upon 
its surface. Most probably during the latter part of this 
period, perhaps for 30 millions of years or longer, completing 
the 114 millions of the first period of our table (p. 225), 
the polar regions may have decreased in temperature by the 


excess of polar radiation cooling the superincumbent vapour 
into rain-clouds of sufficiently low temperature upon precipi- 
tation to allow the commencement of organic life in polar 
regions. This new life, the cause of the beginning of which 
must rest in the Great Unknown, would be afterwards slowly 
distributed in succession further and further from the poles 
as the latitude-temperature fell by the continuous secular 
cooling at greater distance therefrom. The distribution 
under changes of circumstances would cause local variation 
of species outwards from the pole for adaptability to the 
surface-conditions ; but as life- variation is a very slow process 
there would still be strong affinities in the newer animal life 
as it extended, with that which preceded its departure from its 
primitive polar home. 

336. The Archaean rocks are here taken to be the earliest 
chilled superficial rocks, for the greater part projected to the 
surface by plutonic forces and acted upon by the ruling con- 
ditions present ; but if we consider that the same forces are 
still evidently active to a certain extent in producing plutonic 
and volcanic phenomena, we can but take it that the system 
is more or less a continuous one lasting until the present time. 
The outward form of these rocks, particularly the mode of 
crystallization, must certainly vary in time from the differences 
of surface-conditions of the globe, in the heat maintained at 
the surface, the pressure of the atmosphere, the amount of 
aqueous vapour, and the mass of such elements projected to 
or beneath the surface at any period ; but the materials must 
vary very little in chemical composition, coming, as they are 
here suggested to do, from the same universal source of 
exterior nebular deposition. 

337. As we leave the Archsean period, we have new lights 
breaking upon us to guide us to the evidences of past time 
shown in the somewhat systematic evolution of organic life. 
But in this we have to contend with a constant erasure of 
evidences through the activity of the atmospheric phenomena 


within any period, as before stated, which may leave us but 
faint indices of the changes through which the life and 
materials of surface-rocks may have passed before the rocks 
that remain to observation were formed. Thus in the 40,000 
feet or so of old Cambrian and Silurian rocks to be presently 
considered, there is little evidence that we can derive from our 
experience of rock degradation and deposition that these rocks 
were formed directly from Archaean rocks. Indeed we must 
almost conclude from the average fine structure of the Silurians 
that they may have been churned up by atmospheric forces very 
many times during the Archaean period before they arrived at 
the state in which we find them in the Silurians. What we 
appear to possess indices of in rock-texture in many cases, 
is that there were within the Archaean period some special 
periods of deposition, separated possibly by longer periods, 
wherein rocks were deposited and disintegrated many times 
before they formed the strata we find open to observation which 
may have been formed by a special local deposition at a certain 
period. It is also probable that these periods of deposition 
were apart from any uniform conditions and depended greatly 
upon certain special conditions that were active upon the 
globe from the astronomical causes already discussed. 

338. Cambro-Silurian Period. Long before the condensa- 
tion of the Venus-zone previously described, nebulous matter 
condensed interior to the earth's orbit would be floating in 
spiral bands towards this zone, sunward, obscuring a large 
part of the light and heat of the then large nebulous sun, 
so that the obscuration caused by the condensation of the 
Yenus-zone when it was abandoned by the sun might not be 
very sudden ; but as this zone cooled, the obscurity would 
constantly increase up to a certain point of its condensation, 
possibly even so far cooling the earth as to produce a tem- 
porary glacial period about the poles. 

339. At the commencement of the formation of the Yenus- 
zone, the vapour-laden atmosphere would exert a pressure 


much greater than the present. The heavy temperate rains 
that fell at first over the polar areas only, would gradually 
extend over the present temperate regions, by which the 
surface-rocks would be worn off and washed down to the sea- 
shores, and pelagic life spreading originally from the poles 
would inhabit these temperate shores. The shores, by the 
excessive rainfall would be again covered with new deposits, 
brought down from the still elevated Archaean rocks, in which 
the mineral remains of organic life would be buried until 
the sea-bottom extended outward to a great area from the 
coast. These deposits would be again further diffused by 
oceanic currents. The sea would constantly but slowly rise 
by the amount of deposition in proportion to the inland 
denudation, and in a small degree by the loss of vapour in the 
atmosphere, and overflow its former boundaries. 

340. The polar regions, by the effect of constant rainfall, 
would become degraded, as they would be also falling in 
temperature through the obscurity of the sun, until in pro- 
cess of time the cold would be sufficient at the most obscure 
period to cause life to become extinct through wide circum- 
polar regions. At the same time, under diminished heat from 
the sun, marine life, spreading outwards over the then tem- 
perate latitudes from the polar regions, would be supported by 
convection currents from the warmer submarine rocks not 
yet cooled to surface-temperature, so that life would extend 
over the entire temperate and tropical regions of the 

341. In the most obscure part of the period the initial heat 
of the globe, superficially cooled down, would maintain the deep 
waters in the tropical and temperate latitudes upon its surface 
at a fairly equable temperature favourable to marine life. 
Although under these conditions, if by ebb of tides or other 
circumstances such marine life was exposed in restricted areas 
in circumpolar regions without the circulation of warm 
currents, it would be destroyed by the low temperature 


from the little heat given by the obscure sun, particularly in 
the winter solstice. 

Towards the time of extreme obscurity of the Venus-zone, 
copious snowfalls would occur over the polar regions, which 
would press upon the chilled surface with sufficient weight to 
react upon the still semi-fluid siliceous rocks immediately 
underlying the surface-strata, so that these rocks might be 
pressed outwards in under-currents and underflow the surface- 
rocks, causing extensive local and mountainous elevations over 
many regions. These lower heated rocks would also overflow 
the earlier rocks in some districts by exudation due to reaction 
of polar pressures. *it**- 

342. I place the duration of the Cambrians and Silurians, 
some time antecedent to and throughout the first dull period 
given in our table, p. 225, during which time the earth 
was sufficiently cooled down to permit the spread of marine 
life, at 96 millions of years according to our scale. This period 
to the field-geologist may appear relatively short to represent 
the great extent of deposition, which is locally in some districts 
40,000 feet or more. We must, however, in this case consider 
the active state of the atmospheric conditions present, which 
would produce a nearly constant rainfall over a large part of 
the globe, and therefore that it was a period of constant 

343. We have further the probability that the subaqueous 
equatorial regions of the earth continued sufficiently heated 
near the solid surface through all this period to produce a 
constant copious evaporation of the equatorial oceans, so 
that from this cause again the temperate regions would receive 
excessive rainfall, which would flow down to the warm oceans 
carrying mineral matter. The time, again, appears short for 
the evolution of the great quantity and variety of organic 
remains of pelagic life that we find in this period. But here, 
again, we have the probability of life having originated much 
earlier over the cooler polar regions in later Archaean times, 


as before stated, 153, and the possibility at this time, from 
the large disc of the dull nebulous sun subtending an angle, 
as proposed, of 67 degrees, that the small amount of heat 
derived from the sun was so distributed over temperate 
regions that it was particularly conducive to the rapid spread 
of marine life in the constantly subaqueously-warmed seas. 
In fact it was, as the fossil remains indicate, a perfectly 
equable quiescent temperate time, although under very 
constant rainfall except only that the polar regions in the 
most obscure time may have been deeply covered with snow 
and ice, as before stated, to produce about these regions a 
temporary glacial period with contemporary elevation of 
surface-rocks elsewhere upon principles already discussed. 
Daring the publication of this work it has come to my notice 
that the equable condition of Silurian time, as being due to a 
large nebulous sun, has been suggested by M. C. Wolf in his 
able work ' Les Hypotheses Cosmogoniques,' p. 32. 

344. The general evidences of the Cambro-Silurian time 
indicate that comparatively shallow seas prevailed or extensive 
flat shores which abounded with life. The frail shells indicate 
quiescent oceans free from storms and from much tidal action. 
These shores were constantly thickened by new deposition of 
fine sands, mud, or calcareous concretions, consequently to 
remain shallow they must have sunk or the ocean must have 
risen. Most probably both these conditions occurred. The 
recently highly-heated rocks would be slowly shrinking by 
loss of heat in superficial strata ; and at the same time in 
this early age there would be a slow underflow of the still 
viscid lower heated rocks that were moving equator-ward to 
gravitation-equilibrium. At the same time the excess of 
rainfall would be lowering the surface-rocks, and in this degree 
raising the oceanic level as before stated, the whole condition 
causing great depths of shore deposits. The ripple-marks 
which give the evidences of quiescent times, independently 
of the even stratification, were possibly preserved to us by 


occasional light clouds of dry dust blown over them from 
the higher lands during summer. 

345. Devonian Period. When the nebular zone of Venus 
had become broken and had commenced to condense to a 
nebulous planet of large diameter its disc would alternately 
obscure the sun and leave it open. At such a time the 
intensity of the heat and light of the sun would be inter- 
mittent in periods of the synodic inferior conjunctions of 
Venus. Such intermittent periods, by causing periodically 
rapid changes of temperature from mild to intense cold, 
would render the land-surface incapable of maintaining 
either vegetable or animal life, and be also destructive of 
littoral mollusk life. But in the open ocean, where such 
atmospheric changes due to temperature would create great 
disturbance, and thereby increase of the oceanic circulation, 
deep-sea mollusks and fishes of migratory habit could 
survive. It is at this period we have possibly the early 
progressive evolution of distinctly vertebrate fishes, which 
appear to be in a certain degree allied to the later forms of 
crustaceans and to the embryo forms of the future reptiles 
possibly evolved partly, I would suggest, by extension of the 
caudal parts of these early animals, so that certain early 
crustaceans represent the head only of the later fishes. These 
fishes, which came to perfect development in the period 
represented by the Old Red Sandstone in Great Britain, were 
evidently forms adapted to the required conditions of migratory 
habit. They possessed in most cases powerful fins and tails, 
the fins in some cases being possibly adapted to rapid 
surface swimming, such fishes of the entire family of the 
Asterolepidse, as Coccosteus dedpiens, Pterichthys ollongus, 
Cephalaspis Lyelli. Other fishes were evidently adapted to 
deep-sea swimming, as Holoptychius nobilissimus, Osteolepis 
major, and many other forms, all of which were equally 
adapted to migratory life. 

346. After the intermittent obscure period there would be 


a period in which the condensing globe of Venus would 
become neutral in its heating effects, being neither obstructive 
nor auxiliary to the sun's heat. At this period we have all 
the conditions conducive to the spread of vegetable life over 
the land-surface. 

347. From the period of greatest obscurity by the conden- 
sation of the nebular globe of Venus to that of neutrality of 
this planet's effects upon the sun's radiation, would be the time 
of greatest deposition of snow upon the cooler portions of the 
earth during the winter solstice. Through this period the 
earth's crust, being much thinner than at present, would 
be more sensitive to upheaval and intrusion of plutonic 
matter beneath the surface from pressure of the snow at the 
poles bearing down the polar surface rocks. There would 
therefore be evidence of plutonic action, taken in its broadest 
sense, almost synodically during this period, elevating im- 
mense districts of land-areas in steps. 

When Venus had condensed to an incandescent state, 
plutonic and volcanic action would cease and the entire snow 
at the poles would be melted, so that there would be a return 
to purely aqueous conditions with elevation of sea-level. 
The entire period represented by the Devonian in Great 
Britain I assume to have lasted about 60 millions of years, 
extending beyond the third period of our table. 

348. Carboniferous Period. This would commence in the 
intermittent bright fourth period of our table. The tempe- 
rature of the air would be raised at the period of internal 
conjunction of Venus and gradually fall, so that there would 
be intermittent periods of bright sunshine followed by heavy 
rains. This would produce thin intermittent stratification of 
rocks upon the shores and in inland lakes. These lakes would 
overflow at the rainy period and cut incipient river-channels, 
leaving thereby extensive marsh-lands, which would represent 
in series extending coastwards the rivers of the time. When 
the sun became open in the fifth period these marsh-lands, 


owing to the uniform temperature and general quiescence of 
atmospheric conditions, would become crowded with vegetable 
life, and the general conditions would also be conducive to the 
evolution of amphibian life. The Carboniferous period would 
extend throughout the entire fifth open period of about 
306 millions of years until the commencement of the clouding 
of the sun for the commencement of the formation of the 

349. The long period suggested above may be thought to 
exceed that demanded for Carboniferous time by the evidences 
presented by geology. The strata represented appear to be 
only from about 2000 to 13,000 feet where deposits are left. 
We must observe that the deposits are generally very fine, and 
indicate slow intermittent formation frequently of fine wind- 
borne sand without animal life, or layers of fine mud that 
was disintegrated by gentle rains from the rocks into the 
shallow surface pools. These were possibly filled with 
delicate cellular animal and vegetable life, which has left no 
remains. Such light strata were again easily eroded after 
deposition under plutonic elevation, and have no doubt for 
the greater part entirely disappeared, being incorporated 
into later formations. 

350. Permian Period. When the sun was obscured by the 
incipient condensation of the Mercury-zone, the conditions 
given for the first dull period when the Venus-zone was 
formed, which commenced the Cambro-Silurian times, would 
be nearly repeated, but now the terrestrial conditions under 
which it was superimposed would be entirely different. The 
earth's crust would have greatly thickened by cooling, so 
that the tropical ocean-bottom would no longer heat the 
superimposed waters in a sensible degree to cause great 
evaporation over these regions, as in Silurian times, and 
heavy warm rains over polar regions therefrom. The general 
surface-strata would also be cooled down, so that the con- 
ditions of cooling which in the case of the Silurians induced 



the spread of pelagic life would in this latter case cause 
its destruction. The general cooling of the earth .at this 
period by the obscuration of the sun's heat therefore pro- 
bably caused a steady drain of vapours from the surface of 
the ocean, seas, and lakes, which was deposited as snow upon 
the higher land-surface. The lakes on evaporation left saline 
deposits in their beds, combined with fine dust or sand blown 
from the higher rocks disintegrated by frost. The accumu- 
lation of snow over all the present circumpolar temperate 
lands caused, by its pressure upon the still semi-fluid lower 
rocks, the general elevation of lands of the earlier periods 
more distant from the poles, floating these up bodily in 
some districts and by local disturbance with unconformity in 

351. It is during the Permian period as it is represented 
in Great Britain that we find the dying out of palaeozoic 
life. In the early most obscure period possibly the conditions 
were such that life became extinct within the entire present 
temperate regions of the globe, except in the deep waters of 
the ocean. Upon the earth's surface within the tropics alone, 
where the dim sun could effectively diffuse its direct though 
obscure rays, prevalence of life was possible. 

352. In the continuity of life after the dull period we have 
directly the reverse conditions to those that ruled after the 
earlier dull Silurian period. We have life spreading from the 
tropics instead of from the poles. The migratory species that 
could reach the temperate shores were also generally much 
more highly organized and locomotive, and adapted to bring 
in the important factors of animal existence in the new era of 
the Mesozoic period. 

This dull Permian period would break into the period of 
the separation of the Mercury-zone, by which an intermittent 
period would be caused, again producing rapid changes of 
ihin stratification of rocks, not at first conducive to the 
existence of organic life till the condensation of Mercury 


ceased to obscure the sun. What we may possibly define 
as the Permian period lasted nearly through the sixth dull 
period, which, according to our table, p. 225, would be about 
60 millions of years. 

353. Triassic and Rhcetic Periods. This division would 
complete the second dull to bright intermittent period, when 
Mercury was formed as a nebulous planet obscuring the sun 
when in inferior conjunction. The rapid changes produced 
in temperature would cause rapid deposition of rocks and 
greatly restrict the conditions of organic life ; but gradually 
the volume of Mercury would become less before entering 
into the state of an incandescent planet and more calm 
conditions would prevail. The Triassic period would last 
about 47 millions of years, completing the open to dull period. 

354. Jurassic Period. The eighth of our table or the second 
auxiliary period. We again enter into a time particularly 
adapted to the rapid evolution and spread of pelagic organic 
life. The sea-shores by the intermittent rainy conditions 
due to temperature changes, yet always warm, became 
covered with fine mud from the land-areas, bringing down 
with it abundance of support for a suitable class of littoral 
life, carrying at the same time into the deeper oceans carbonate 
of lime in solution available for all forms of life depending 
upon it. This period would extend throughout the eighth 
division of our table, for about 37 millions of years. 

355. Cretaceous and Eocene Periods onwards to tlie present 
Time. These may have been partly contemporary if we 
imagine that the Chalk was a formation more distant from 
the ancient shores, which was continuously over-deposited in 
some areas, while elsewhere it was covered by argillaceous 
rocks, as the land-surface was degraded. This is taken to 
represent the ninth period of our table. This period, under 
constant diminution of the sun's disc, would be generally 
equable, producing only gentle deposition of rocks locally 
in drainage-basins or in the distant ocean-bottom. It would 


otherwise be only subject to the set of changes that were 
due to the condition of the sun at the critical temperatures 
of its former nebulous surroundings, which may have pro- 
duced occasionally more or less obscure periods, to which 
we may add the changes due to eccentricity of orbit and 
variation of axis, which have produced great changes in 
climate. This long period drifts us into the present time, 
when the sun's disc has become very small relatively to what 
it was in the past and of intense light and heat. 

The period set aside by our table to include the Cretaceous 
and Tertiary periods is 532 millions of years. This may 
appear long for the number of superficial changes that are 
evident upon the surface from the amount of mineral matter 
deposited. In such a uniform period the variations were 
probably all local and for the most part intermittent through 
the periodical astronomical changes, so that land was alter- 
nately deposited and degraded many times without producing 
any great entire depth of strata *. 

356. With regard to life, our only true index of time in 
this period, we may conclude that where there is a general 
constancy of like conditions there is little reason for change 
for adaptibility to the circumstances present, particularly if 
the organism is elevated to a condition of migratory instinct 
for accommodation to the seasons. Therefore, seeing that 
the changes have been great, the evolution period must have 
been immense to have produced the variation in forms which 
we know to have occurred in this period, more particularly in 
that of the elevation of the scale of the higher mammals. 
The changes within the tertiary period present wonderful 
variations in the structure of the mammalia to give the number 
of species we at present possess without consideration of the 
number extinct. There is one mark, however, of continuous 
progress throughout all this period the brain constantly 

* Appendix C. 


grows larger in relation to the bulk of the animal. This is 
possibly a very slow continuous progressive feature, which 
may give us some idea of the enormous extent of time 
embraced within this period. 

357. We may generally postulate from geological evidence 
that in the long Tertiary period we have had present the 
conditions proposed by Lyell, in his ' Principles of Geology,' 
for the entire system of deposition of surface rocks, wherein 
every deposit is assumed to depend upon local circumstances, 
and the rocks of the past to be simply disintegrated and 
redeposited without any important change in the cosmic 

358. In the above discussion of periods the effects of 
changes of eccentricity of orbit and variations of the obliquity 
of ecliptic have only just been mentioned. These variations, 
there is no doubt, produced marked effects and caused 
changes of climate locally which have produced minor 
divisions in stratified rocks. This subject has been ably 
discussed by others, and probably in some cases its import- 
ance has been much exaggerated *. 

359. There are also many conditions that have materially 
affected the contemporary stratification, which depends greatly 
upon the sea-level of the period, some of these have been 
considered, but they still leave generally a wide field for 
future investigations. The most important is the study of 
the equilibrium of the earth's mass as a gravitation system 
in rotation under the superficial changes of surface rocks 
that are evident in periods of the past ; thus, for instance, 
the greater or less elevation of ice at the South Pole would 
disturb the gravitation centre. A past reaction of ice at the 
North Pole, which probably, upon conditions proposed above, 
was the direct cause of elevation of the entire plateau of 

* Lyell's f Principles of Geology,' vol. i. p. 272 ; Author's paper, British 
Association Reports, 1884, p. 723. 


Central Asia. This would again react on the earth's equi- 
librium. Both of these systems would tend to lower the 
sea-level of Great Britain; but the subject is too large to be 
even sketched in the present treatise. 

360. Glacial Period. As regards this period, which neces- 
sarily comes within the ninth division, this, as a consequence 
of the natural diminution of the sun's volume, has been 
discussed in Chapter XIII. My early reflections upon this 
subject, when strongly imbued with the Huttonian principles 
developed by Lyell, led me to think that there were probably 
indices of sufficient change under local conditions with slight 
variations of elevation of land-areas to account for this epoch 
in Europe. These conclusions were mainly based upon my 
idea that the movements of aerial and oceanic currents are 
caused by the expansion of the atmosphere, and in less degree 
of the oceanic surface, by the heat of the sun in its apparent 
diurnal motion through the tropics driving before it an expan- 
sion-wave of air and water *. This as a systematic motion, 
according to my theory, would produce whirls or cyclones 
lateral to the tropics, the position of which would depend upon 
the resistance of the coast-lines. Under these conditions the 
resistances that locate the North- Atlantic whirl are the coasts 
of North America. The peculiar conformation of this 
coast at the present time deflects the North-Atlantic 
whirl into a bi-whirl, of which the northern part that 
is, the Gulf Stream crosses the Atlantic, passes along 
the coast of Norway, enters the Arctic Circle, and is de- 
flected back to its origin after passing along the coast of 
Greenland, which it leaves glaciated by the cold Arctic 
current. This direction of rotation is quite abnormal to that 
of any other lateral tropical whirl. It is clear, as stated in 
my paper read before the British Association in 1885 f, that if 

* Fluids/ p. 626. 

t Brit. Assoc, Reports, 1885, p. 1020. 


the lower central lands of North America were only sunk some 
300 feet from the mouth of the Mississippi through to Hudson's 
Bay, a result readily produced by cosmic causes already 
discussed, this whirl would then take its normal course, as 
other lateral whirls in the South Atlantic, Pacific, and Indian 
Oceans. Under these conditions Northern Greenland would 
be placed in the tropical current and enjoy a very temperate 
climate, and the returning Arctic oceanic and aerial current 
would bathe the coasts of Western Europe, leaving icebergs 
on the coast with inland glaciation, as at present in Green- 
land, This would include also the glaciation of the northern 
part of Great Britain. 

361. To account for the glaciation of Northern America 
upon like principles to the above, we should require the mean 
temperature of the Northern Pacific Ocean to be sufficiently 
high for its whirl deflection to keep the Behring Sea open, so 
that the Northern Pacific whirl could continue its direct 
motion in open water north of Alaska, bringing the return 
Arctic aerial currents into the valley of the Mackenzie River 
and through the Great Bear and Great Slave Lakes into 
the valley of the Missouri, laden with sufficient moisture to 
produce a contemporary glacial period in the northern parts 
of the South-western States and distribution by currents 
through the lateral valleys. 

362. The considerations which have made me somewhat 
modify this idea, without change of the principles suggested 
so far as they are active, were due to a more attentive study of 
American geology. The glaciation of Arctic North America 
appears to have been greater than these principles would 
entail. This to my mind is seen most particularly, according 
to my theory, in the evidence of the great outflow of basalt 
in the region of the Snake Biver, Idaho,- within tertiary 
times. To produce this great outflow of lower heated rocks 
upon principles herein discussed, there must have been very 
great elevation of ice, most probably about the North Pole. 


I assume that a pressure system of ice at the poles would 
react throughout the entire lower viscous rocks; hut as these 
rocks are assumed to be supported by flotation upon the- 
denser metallic nucleus, ihe reaction by protrusion to restore 
gravitation-equilibrium at a distance from either pole would 
be much more frictional than that from a nearer pole. The 
mass of basalt protruded in Idaho possibly equals the entire 
mass of land above the oceanic surface in Great Britain *. 
There was also contemporary elevation of the great plateau 
of Central Asia. There may therefore have been an ice-cap 
in the north, somewhat equivalent to that of the Antarctic 
Circle. The cold necessary to produce such an ice-cap we 
can scarcely imagine to have existed at the present mean 
temperature of the globe. It is therefore more consistent to 
assume that the effective radiation of the sun was diminished 
for a period, probably, as I have proposed, by clouding in the 
condensation of nebular matter at its critical temperature, p. 74. 
This cannot, however, destroy the evidence of glaciation. being 
at any time local in intensity, and the marine shells in the 
glacial clays indicate an open ocean poleward f. There is 
said to be no evidence of glaciation in the great plains of 
Siberia, and from my own observation there has been none 
in the west of Norway ; for instance, upon the granitic and 
gneiss rocks of the Lofoten Isles which retain the sharp 
pointed outlines of ancient rocks that have been subject to 
weathering only throughout long geological periods J. 

For great elevation of circumpolar land or massive outflow 
of basalt at any period, the extreme cold of the previous 
period may have produced great rigidity in the ice-system 
and contiguous rocks. The reaction of such a system would 
cause the more distant parts of the earth's crust to give way 

* Geikie, ' Geology/ p. 257. 

t ' Acadian Geology/ Sir J. W. Dawson, p. 65. 

J Author's paper, Geol. Soc. Proc., Feb. 23, 1887. 


paroxysmally and after fracture to continue the effects of the 
reaction until the polar pressure reached nearly its point of 
equilibrium. In this manner reactions of ice-pressures would 
be periodical upon the surface system of rocks. 

363. Future Period. By the continuity of the conditions 
which now rule, the sun will not probably become a much 
brighter incandescent body than it is, and it will decrease 
in volume and ultimately in incandescence. The ice-caps 
which cover the Antarctic and probably the Arctic pole 
must therefore grow more extensive with the decrease of 
solar heat ; and although increase of weight of ice may 
cause deflection of the crust and distribution of pressure 
upon the interior highly heated rocks, which pressures 
may react in producing plutonic and volcanic phenomena, 
still, with each such displacement the crust will become 
thicker, more rigid, and more resistant, and thereby vol- 
canic and plutonic action to overcome the resistance will be 
more paroxysmal. The oceans also, by the condensation of 
evaporated water and its deposition in the form of -ice at 
the poles, will become of less depth, so that land-areas will 
increase in aerial surface. 

364. The periodic process would therefore appear to be a 
general decrease of temperature accompanied by apparent 
elevation and spread of continental lands and decrease of 
oceanic area, so that the temperate inhabitable globe would 
increase over tropical areas for a long period in greater ratio, 
possibly, than the circumpolar areas and would cease to be 
sufficiently temperate for the existence of organic life. 

At a later period the evaporation from the tropical oceans 
would be less, and deposition in circumpolar areas less, and 
the earth's crust more rigid, so that the sea would diminish 
less by this cause; but at the same time the mountainous lands 
surrounding the great oceans would have the snow-line 
lowered, so that a part of the evaporation from the oceans would 
fall upon the adjacent lands instead of drifting poleward, 


and the lower shelving shores of the much reduced tropical 
oceans would hecome habitable lands. 

At a still later period the tropical ocean-beds would be 
drained by evaporation in clouds drifting over to the shores, 
which for the most part would never return to them in the 
frozen river-streams. The earth would then possess three or 
four oceanic areas adapted to life only the lower beds of the 
Pacific, Atlantic, and Indian Oceans. 

Later when the sun presented only a dull red disc, appear- 
ing to move daily across the starry vault, the lands repre- 
senting the deep ocean-beds would be frozen and life gradually 
become extinct. 

365. The entire cloud-drainage of the great oceans and 
the snow-clad mountainous lands surrounding them would 
release the pressure upon the deep ocean-beds in proportion 
to the increase of the weight of snow upon the mountains. 
The earth would probably still be sufficiently yielding in its 
interior to admit of a certain amount of reaction by distribu- 
tion of surface-pressures, by which the ocean-beds would be 
elevated to restore partial gravitation-equilibrium. This 
effect would probably be produced by distribution of small 
volcanoes over the former lower oceanic surface, and, as the 
surface would be shrinking slightly by loss of temperature, 
there would occur also paroxysmal upheavals in localities 
where, through the tension and plutonic pressure below the 
surface, resistance would be overcome. This would leave 
earthquake-fissures as a permanent surface-feature. 

366. It is possible that in one of the warmer tropical 
valleys running east to west formed from one of the above- 
described fissures in the deepest bed of the Pacific, and near 
to some residual thermal springs, the last individual, of the 
latest evolved form of humanity, may die of hunger and close 
for ever the records of science attained upon our globe. 

Whether extinction of life will occur within the short 
period of 15 millions of years, as suggested by the theory of 


Helmholtz for the sun to decrease to the density of the earth, 
can scarcely be suggested. Whether we know the sun's 
specific heat or the law of dispensation of solar heat into 
space is doubtful ; but the probability is that world-life will be 
longer than this, if we can accept the conditions which I have 
reserved for discussion in Appendix A as my theory on this 
point may be thought not to be a necessary part of our 

[ 252 ] 


I MAY offer as a pure hypothesis that the energy of a light- 
and heat-giving system may not be so rapidly dissipated as we 
know it to be by experiment unless it meet in radiation with 
another material body as a recipient. In fact, that for the 
rapid diffusion of light and heat through cosmic space there 
must be one or more motive couples, just in the same way 
as this is necessary for the action of gravitation-energy, only 
that light and heat bear reference to surface only, not to mass. 
Upon this hypothesis light and heat, as possibly gravitation, 
may be considered in certain cases as phenomena of induction, 
and in action, in a certain degree as regards the sun, 
resemble the discharge from one excited conductor to another 
through an insulated medium. The intensity of propagation 
of forces from the sun's globular mass being assumed equal 
to that of the discharge of electricity from a point into a 
space of direct insulation that is, insulation from general 
diffusion, so that the induction to another body if present 
falls in direct line only, with the loss only of a limited 
amount by free radiation. In this case the form of force is 
nevertheless that of light or heat, not of electricity, from 
which it may vary in any motive degree. A theory upon 
these lines, to which I have devoted some years of study, 
but cannot extend here, except to state the principle, would 
account for there being snow-caps about the poles of Mars, 
whereas, from his distance being taken inversely as the square 
in comparison with that of the earth, the amount of sun-heat 


capable of reaching the surface of this planet would leave it 
wholly frozen. The intensity of the light of Jupiter and 
Saturn also far exceeds that due to the reflection of the sun's 
light as deduced from calculation of uniform radiation only. 
Upon these principles, if they hold, it may be suggested that 
we should materially conserve solar energy, for however 
closely light induction couples may fill up the radiants about 
the sun, the open interspaces where there would be less loss 
of energy must be of immensely greater area that is, as 
the space to the mass. Further, the sun would receive as much 
light- energy from another star as it would impart to this 
star, so that the radiant force dissipated would be principally 
upon the planets that is, upon bodies cooler than itself. 
Such a principle of dispensation of heat and light would 
prolong the sun's future life to many times the period esti- 
mated by Helmholtz. 

To separately define the forces active upon a planet in 
relation to the sun in factors of induction according to this 
hypothesis, I would suggest : 

Gravitation. Mass Induction, producing a tendency to draw 
two bodies together whose equilibrium is only satisfied when 
the attraction produces a unit globular mass. 

Heat. Conductive Induction, by which a certain depth 
of mass is affected in diminishing ratio in intensity by some 
geometrical power. This force is probably, if taken per se, 

Light. Surface Induction, affecting the surface molecules 
only to a limited depth, except in special or dioptric bodies, 
through which it passes to other bodies by a system of radial 
conduction to the inductive body beyond. This force is also 
per se probably repulsive. 

As I may never publish my researches on this subject, I 
may say that my idea of light is that its induction renders 
all bodies under its influence luminous to a certain extent. 
That this self-luminosity induces a like secondary luminosity 


in another body or a similar effect upon the retina of the eye 
or a sensitive film in a camera. That this luminous action is 
probably caused by rotating the surface molecules so as to 
cause them to present the surface which affinity in the dark 
draws inwards to an outward position. A complete revolu- 
tion being necessary for white, and a partial revolution for 
colour. But this last stated idea is immaterial, the self- 
luminosity is material. To mention one of my first 
experiments. I made in 1873 a drawer 6 by 6 inches, of one 
inch in depth, very carefully fitted in a velvet-lined frame to 
exclude light. The inside of the drawer could be exposed to 
sunshine while it was closed and drawn into a dark room 
when required. I attached two spiral springs to the front of 
the drawer and a catch to keep it closed when out in the sun- 
shine, so that when the catch was released the drawer came 
instantly into the dark room. I tried many experiments ; the 
first was that of writing my name boldly in Indian ink upon 
a piece of white paper. After this had been a minute or less 
in the sunshine, and was then drawn into the dark room, I 
found that I could read it easily for a short time in the dark ; 
therefore, I conclude it retained a part of its luminosity, or 
at least, if it was luminous in the dark, it must have been also 
luminous in the sunlight, so that its perception to the eye could 
not have been entirely from reflected sunlight as generally 
assumed. It appears to me, therefore, that we are bound 
to admit induced luminosity as a factor of visibility. These 
effects, as phenomena ascribed to phosphorescence, are well 
known, and have been investigated most ably by Becquerel ; 
but my idea of them is that they are not, as assumed, simply 
phenomena of phosphorescence, but of induced luminosity, 
and that they are universal for all light-giving bodies, that 
is, for all visible bodies or such as are not dioptric, or for 
black bodies, if any exist, or so far as they exist. In this 
hypothesis it is not necessary to suppose that a body may 
retain its induced luminosity for an instant in the dark. 


Heat conductors, that is metallic bodies, possess no such 
power of retention, the surface molecule being assumed in 
this case to be sensitive instantly to light and heat influences. 
This does not affect the laws that govern the action of light, 
such as the reflective properties under the condition that a 
body may receive luminous induction in one direction and 
dispense it at coincident equal angles, or the refractive 
properties of dioptric bodies, by which the inducing rays 
are bent, only that in this last case the inductive body 
is behind the dioptric, which acts only as a conductor thereto 
in the same manner as a metal wire does to electricity, but 
following its own laws. The direction of the light force of 
induction is otherwise always in direct line. My experiments 
upon these hypotheses were made in 1872-4, and I have 
discussed some of them with my friends. Some of these ideas 
appear to have occurred to a correspondent of the ' English 
Mechanic,' T. W. B., last year, 1894, and, so far as that publi- 
cation goes, I acknowledge the priority if it is of any value. 
In the multiplicity of thoughts, by reason of our similarity 
of organism, some of our ideas must be like those of other 

[ 256 ] 


THE generally accepted theory of land-formation is that 
which was proposed or maintained by the late "Robert Mallet 
in a paper upon " The amount of Energy developed by the 
Secular Cooling of the Earth," contained in two papers, over 
100 pages, in the Phil. Trans. 1874-5. According to these 
papers the amount of heat lost from the initial temperature of 
the earth will represent the force of its contraction. The 
amount of this energy is presumed to be made evident in 
compression of the superficial strata causing the elevation, 
inclination, and crumpling of the strata and the entire 
volcanic phenomena. The data, upon which the arguments 
of these papers rest, are assumed to be taken from calculations 
of Elie de Beaumont, Forbes, and Lord Kelvin, who estimate 
the heat lost by the earth to be equal to the melting of a 
plate of ice, respectively of 0'0065, 0*007, and 0-0085 milli- 
metres annually. From these data it is stated that from 575 
to 777 cubic miles of ice melted annually would represent the 
loss of heat. By going over the calculations in this paper, I 
was able to point out a considerable error in it, sufficient to 
upset the whole contraction theory upon the lines laid down 
by Mr. Mallet. After writing to Sir George Stokes, then 
Secretary to the Royal Society, who clearly saw the acci- 
dental error, I read a paper upon it before the Geological 
Society * in June 1884, showing that the contraction from the 
data given was only about one cubic mile annually, that is, 

* Quart. Journ. Geol. Soc. vol. xl. (Proc.) p. 67. 


from -7937 to 1'03S7 mile. The principal authority for the 
data given was Lord Kelvin ; and as I could not find any 
reference to the subject in his papers, Sir George Stokes 
kindly wrote to Lord Kelvin for me about this, and found 
that the assertion was altogether a mistake. Lord Kelvin 
never made such a calculation, therefore this theory, sup- 
ported upon the evidence of compression of surface-strata, is 
generally without foundation in fact. I think, moreover, that 
the contraction theory is quite opposed to observation of actual 
rocks, the joints of which are generally open below the surface, 
and show the effects of pressure from beneath producing a 
tensile strain upon the surface to form the open joints. These 
joints are often filled with basalt from intrusion of the under- 
lying magma, showing more directly evidence of the outward 
pressure of heated liquid plutonic matter. 

[ 258 ] 


IF the general theory of this work is accepted at some future 
time, a more experienced practical geologist than myself 
may shift the divisions in the rock-series that I have adopted 
to make them more exactly agree with the periodic conditions 
proposed. To do this perfectly would require refined geolo- 
gical observation, as the astronomical changes herein defined 
could not have been generally abrupt, so as to produce very 
distinct divisions in stratifying rocks. Further, there must 
be superimposed upon the greater astronomical changes herein 
suggested, the minor influences of variation of eccentricity 
of orbit and change of obliquity of axis, which would produce 
variation of deposition although possibly not to the extent 
proposed by Croll and Lyell. Some objection, for instance, 
may be made to my grouping the cretaceous with the tertiary 
in one long period, wherein the chalk formation is at least 
very distinct, and locally no doubt, if taken in vertical series, 
the more ancient. In this case we may consider the chalk to 
be a deep oceanic formation that is still in progress, a theory 
generally accepted since the ' Challenger ' Expedition. I 
think a system of contemporary stratification of the various 
kinds of sediment distinguished by special chemical elements 
must have been general throughout all time, as we have 
only one set of such elements largely to deal with upon the 
surface of the globe, however much they may have been 
churned up or sorted out by local prevailing conditions. 
Upon this suggestion we could at no period have had one 


general system of deposition prevailing either of silica, 
alumina, or calcic-carbonate, in other than local areas. The 

general scheme of deposition in quiescent times and undis- 
turbed by oceanic currents may be shown diagrammatically 
by the figure above, which may represent, say 200 miles 
of, deposition from a coast of the ancient rock-surface of a 
certain period : O, the oceanic surface ; a line from the 
ancient rock to a point F the surface of the newly-formed 
rocks, where a band of flints occur in the chalk from organic 
deposition at a certain distance from the coast. Then of the 
produce of the disintegrated rocks, the coarser materials 
would rest at B ; the broken masses of silica or sand at S ; the 
lighter mud or clay at C ; the perfectly soluble carbonate of 
lime and silica at CH, where it would be generally absorbed by 
organic life. This system would in all cases form sets of rocks 
and go on continuously over areas of surface-drainage carrying 
the disintegrated rocks if undisturbed by oceanic currents or 
tidal action, and could in the past only be arrested by such 
great astronomical changes as herein proposed. These greater 
changes cannot occur again, so that the present period may 
be geologically indefinitely extended for the time the ocean 
retains its liquidity. 






This book is due on the last date stamped below, or 

on the date to which renewed. 
Renewed books are subject to immediate recall. 

JUG 919C9I 



FEB14 1969 88 


FEB V69-7P 



^S 6 ' uwW^Sg-i.